
import pandas as pd
import numpy as np

import statsmodels.api as sm
import statsmodels.formula.api as smf
import scipy.stats as stats
from statsmodels.stats.multitest import multipletests

import tqdm

import matplotlib.pyplot as plt
import seaborn as sns

from skimage import filters, morphology, measure, color
from skimage.measure import regionprops
from tifffile import imread, imsave

import glob

from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler

from skimage import measure

from scipy.spatial.distance import pdist

import random
random.seed(42)

plt.rcParams['pdf.fonttype'] = 'truetype'
plt.rcParams['font.size'] = 20
plt.style.use('seaborn-v0_8-ticks')


color_palette = {'DMSO' : '#4d4d4d',
                 'ARYL_AG':'#F8766D', 'ARYL_INH':"#F8766D50",
                 'ANDR_AG':'#fccb17', 'ANDR_INH':'#C49A0050',  
                 "ESTR_AG": '#53B400', "ESTR_INH": '#53B40050', 
                 'GC_AG' : '#00C094', 'GC_INH' : '#00C09450',
                 'LX_AG' : '#00B6EB', 'LX_INH' : '#00B6EB50', 
                 'RET_AG' : '#A58AFF', 'RET_INH' : '#A58AFF50', 
                 'THYR_AG' : '#FB61D7', 'THYR_INH' : '#FB61D750'
}

import matplotlib.colors as mcolors

def hex_to_rgba(hex_color):
    """Convert hex color to RGBA. Supports 8-digit hex with alpha."""
    hex_color = hex_color.lstrip('#')
    if len(hex_color) == 6:
        return mcolors.to_rgba('#' + hex_color)
    elif len(hex_color) == 8:
        rgb = hex_color[:6]
        alpha = int(hex_color[6:], 16) / 255
        return (*mcolors.to_rgb('#' + rgb), alpha)
    else:
        raise ValueError(f"Invalid hex color: #{hex_color}")

def hex_to_rgba_string(hex_color):
    """Convert hex color to rgba() string for Seaborn."""
    hex_color = hex_color.lstrip('#')
    if len(hex_color) == 6:
        r, g, b = tuple(int(hex_color[i:i+2], 16) for i in (0, 2, 4))
        return f'rgba({r}, {g}, {b}, 1.0)'
    elif len(hex_color) == 8:
        r, g, b = tuple(int(hex_color[i:i+2], 16) for i in (0, 2, 4))
        a = int(hex_color[6:], 16) / 255
        return f'rgba({r}, {g}, {b}, {a:.3f})'
    else:
        raise ValueError(f"Invalid hex color: #{hex_color}")
    
# Convert all to RGBA
color_palette_rgba = {k: hex_to_rgba(v) for k, v in color_palette.items()}
color_palette_rgba_str = {k: hex_to_rgba_string(v) for k, v in color_palette.items()}

color_palette_expanded = {'DMSO_CTL08A': '#cacaca',
 'DMSO_CTL04E': '#4d4d4d',
 'ARYL_AG_CTL08A': '#fdd6d3',
 'ARYL_AG_CTL04E': '#F8766D',
 'ARYL_INH_CTL08A': '#fdd6d3',
 'ARYL_INH_CTL04E': '#F8766D50',
 'ANDR_AG_CTL08A': '#feefb9',
 'ANDR_AG_CTL04E': '#fccb17',
 'ANDR_INH_CTL08A': '#ede1b250',
 'ANDR_INH_CTL04E': '#C49A0050',
 'ESTR_AG_CTL08A': '#cbe8b2',
 'ESTR_AG_CTL04E': '#53B400',
 'ESTR_INH_CTL08A': '#cbe8b250',
 'ESTR_INH_CTL04E': '#53B40050',
 'GC_AG_CTL08A': '#b2ecdf',
 'GC_AG_CTL04E': '#00C094',
 'GC_INH_CTL08A': '#b2ecdf50',
 'GC_INH_CTL04E': '#00C09450',
 'LX_AG_CTL08A': '#b2e9f9',
 'LX_AG_CTL04E': '#00B6EB',
 'LX_INH_CTL08A': '#b2e9f950',
 'LX_INH_CTL04E': '#00B6EB50',
 'RET_AG_CTL08A': '#e4dcff',
 'RET_AG_CTL04E': '#A58AFF',
 'RET_INH_CTL08A': '#e4dcff50',
 'RET_INH_CTL04E': '#A58AFF50',
 'THYR_AG_CTL08A': '#fed0f3',
 'THYR_AG_CTL04E': '#FB61D7',
 'THYR_INH_CTL08A': '#fed0f350',
 'THYR_INH_CTL04E': '#FB61D750'}

line_palette = {
    'CTL08A': "#99cfd670",  # azzurro pastello
    'CTL04E': "#f8c88a70",  # arancio pastello
}


labels = list(color_palette.keys())
handles = [plt.Line2D([0], [0], color=color_palette[label], lw=8) for label in labels]

fig, ax = plt.subplots(figsize=(8, 2))
legend = ax.legend(
    handles, labels, loc='center', ncol=int(np.ceil(len(labels)/2)), frameon=False,
    bbox_to_anchor=(0.5, 0.5)
)
ax.axis('off')
plt.tight_layout()

plt.savefig('../../../../endpoints_figures/color_palette_legend.pdf', dpi=300, bbox_inches='tight')

/tmp/ipykernel_1121090/191618692.py:10: UserWarning: Tight layout not applied. The left and right margins cannot be made large enough to accommodate all axes decorations.
  plt.tight_layout()


blank_mix1 = glob.glob('/group/testa/Project/EndPoints/TPSSU/analysis/single_tif_scanregion/*0136*.tif')
blank_mix1 = [f.split('/')[-1].split('.')[0] for f in blank_mix1]
blank_mix2 = glob.glob('/group/testa/Project/EndPoints/TPSSU/analysis/single_tif_scanregion/*0145*.tif')
blank_mix2 = [f.split('/')[-1].split('.')[0] for f in blank_mix2]


check_qc_ch1 = pd.read_csv('../../preprocessing/tables/Treshold_QC_check - Channel1_no_dup.csv')
check_qc_ch2 = pd.read_csv('../../preprocessing/tables/Treshold_QC_check - Channel2_no_dup.csv')


check_qc_ch1['general QC (keep or not)'] = check_qc_ch1['general QC (keep or not)'].str.upper()
check_qc_ch1['general QC (keep or not)'] = check_qc_ch1['general QC (keep or not)'].replace({'BETTER NOT': 'LIMIT', 'MAYBE': 'LIMIT', 'YES?': 'LIMIT', 'SUS': 'LIMIT', 'NOT SURE': 'LIMIT', 'YES ': 'YES', 'YES - 2 CBO': 'YES'})
check_qc_ch1['general QC (keep or not)'].value_counts().index

Index(['YES', 'LIMIT', 'NO'], dtype='object', name='general QC (keep or not)')


check_qc_ch1['Reference Image Report'] = check_qc_ch1['Reference Image Report'].str.replace('.pdf', '')


check_qc_ch1['Reference Image Report'].value_counts()

Reference Image Report
20240424_manuel_lessi_0148_ScanRegion1_rep_1_channel_1    1
20240424_manuel_lessi_0145_ScanRegion2_rep_2_channel_1    1
20240424_manuel_lessi_0145_ScanRegion3_rep_0_channel_1    1
20240424_manuel_lessi_0145_ScanRegion3_rep_1_channel_1    1
20240424_manuel_lessi_0145_ScanRegion3_rep_2_channel_1    1
                                                         ..
20240424_manuel_lessi_0136_ScanRegion2_rep_1_channel_1    1
20240424_manuel_lessi_0136_ScanRegion2_rep_2_channel_1    1
20240424_manuel_lessi_0136_ScanRegion3_rep_0_channel_1    1
20240424_manuel_lessi_0136_ScanRegion3_rep_1_channel_1    1
20240424_manuel_lessi_0136_ScanRegion3_rep_2_channel_1    1
Name: count, Length: 153, dtype: int64


check_qc_ch2['general QC (keep or not)'] = check_qc_ch2['general QC (keep or not)'].str.upper()
check_qc_ch2['general QC (keep or not)'] = check_qc_ch2['general QC (keep or not)'].replace({'NOT SURE': 'LIMIT', '?': 'LIMIT', 'YES ': 'YES'})
check_qc_ch2['general QC (keep or not)'].value_counts().index

Index(['YES', 'NO', 'LIMIT'], dtype='object', name='general QC (keep or not)')


check_qc_ch2['Reference Image Report'] = check_qc_ch2['Reference Image Report'].str.replace('.pdf', '')


check_qc_ch1['Reference Image Report'].value_counts()

Reference Image Report
20240424_manuel_lessi_0148_ScanRegion1_rep_1_channel_1    1
20240424_manuel_lessi_0145_ScanRegion2_rep_2_channel_1    1
20240424_manuel_lessi_0145_ScanRegion3_rep_0_channel_1    1
20240424_manuel_lessi_0145_ScanRegion3_rep_1_channel_1    1
20240424_manuel_lessi_0145_ScanRegion3_rep_2_channel_1    1
                                                         ..
20240424_manuel_lessi_0136_ScanRegion2_rep_1_channel_1    1
20240424_manuel_lessi_0136_ScanRegion2_rep_2_channel_1    1
20240424_manuel_lessi_0136_ScanRegion3_rep_0_channel_1    1
20240424_manuel_lessi_0136_ScanRegion3_rep_1_channel_1    1
20240424_manuel_lessi_0136_ScanRegion3_rep_2_channel_1    1
Name: count, Length: 153, dtype: int64


check_qc_ch1['ImageName_ScanRegion_Rep'] = check_qc_ch1['Reference Image Report'].apply(lambda x: '_'.join(x.split('_')[:7]))
check_qc_ch2['ImageName_ScanRegion_Rep'] = check_qc_ch2['Reference Image Report'].apply(lambda x: '_'.join(x.split('_')[:7]))


to_keep_ch1 = check_qc_ch1[check_qc_ch1['general QC (keep or not)'].isin(['YES', 'LIMIT'])]
to_keep_ch2 = check_qc_ch2[check_qc_ch2['general QC (keep or not)'].isin(['YES', 'LIMIT'])]

len(set(to_keep_ch1['ImageName_ScanRegion_Rep']).intersection(set(to_keep_ch2['ImageName_ScanRegion_Rep'])))

100


len(check_qc_ch1['ImageName_ScanRegion_Rep'])

153


len(check_qc_ch2['ImageName_ScanRegion_Rep'])

275


good_tissue_to_keep_limit = list(set(check_qc_ch1[check_qc_ch1['general QC (keep or not)'].isin(['YES', 'LIMIT'])]['ImageName_ScanRegion_Rep']).union(set(check_qc_ch2[check_qc_ch2['general QC (keep or not)'].isin(['YES', 'LIMIT'])]['ImageName_ScanRegion_Rep'])))


len(good_tissue_to_keep_limit)

245


good_tissue_to_keep_yes = list(set(check_qc_ch1[check_qc_ch1['general QC (keep or not)'].isin(['YES'])]['ImageName_ScanRegion_Rep']).union(set(check_qc_ch2[check_qc_ch2['general QC (keep or not)'].isin(['YES'])]['ImageName_ScanRegion_Rep'])))


len(good_tissue_to_keep_yes)

211


sample_sheet  = pd.read_csv('../../preprocessing/tables/ImageAnalysisSampleSheet.csv')
sample_sheet


sample_sheet['ImageName'] = sample_sheet['Path'].apply(lambda x: x.split('\\')[-1].split('.')[0])
sample_sheet['ImageName_ScanRegion'] = sample_sheet['ImageName'] + '_' + sample_sheet['Scan Area']
sample_sheet['Condition_Line'] = sample_sheet['Condition'] + '_' + sample_sheet['Line ']
sample_sheet


sample_dict= {i:j for i, j in zip(sample_sheet['ImageName_ScanRegion'], sample_sheet['Condition_Line'])}


mix_1_names = sample_sheet[sample_sheet['MIX'] == 1]['ImageName'].tolist()
mix_2_names = sample_sheet[sample_sheet['MIX'] == 2]['ImageName'].tolist()


hormonal_conditions = ['ESTR_AG', 'ESTR_INH', 'ANDR_AG', 'ANDR_INH', 'ARYL_INH',
       'ARYL_AG', 'LX_AG', 'GC_AG', 'LX_INH', 'GC_INH', 'THYR_INH',
       'THYR_AG', 'DMSO', 'RET_AG',
       'RET_INH']
hormonal_conditions.remove('DMSO')
hormonal_conditions.sort()
hormonal_conditions = ['DMSO'] + hormonal_conditions


measurements_files = glob.glob("/group/testa/Project/EndPoints/TPSSU/analysis/measurements/*.csv")
measurements_files_mix_1 = [f for f in measurements_files if any(name in f for name in mix_1_names)]
measurements_files_mix_2 = [f for f in measurements_files if any(name in f for name in mix_2_names)]


mask_file = pd.read_csv("/group/testa/Project/EndPoints/TPSSU/analysis/cyto_mask_measures.csv")
mask_file


mask_file['OriginalImageName'] = mask_file['Unnamed: 0'].apply(lambda x: '_'.join(x.split('_')[0:4]))
mask_file['ImageName_ScanRegion'] = mask_file['Unnamed: 0'].apply(lambda x: '_'.join(x.split('_')[0:5]))
mask_file['ImageName_ScanRegion_Rep'] = mask_file['Unnamed: 0'].apply(lambda x: '_'.join(x.split('_')[0:7]))
mask_file


mask_file_df_mix_1 = mask_file[mask_file.OriginalImageName.isin(mix_1_names)]
mask_file_df_mix_1


mask_file_df_mix_2 = mask_file[mask_file.OriginalImageName.isin(mix_2_names)]
mask_file_df_mix_2


def filter_nuclei(df):
    min_quantile = df['area'].quantile(0.01)
    max_quantile = df['area'].quantile(0.99)
    df = df[(df['area'] > min_quantile) & (df['area'] < max_quantile)]
    return df


tot_df_mix1_ch1 = pd.read_csv("/group/testa/Project/EndPoints/TPSSU/analysis/measurements_mix_1_ch1_CTIP2.csv")
tot_df_mix2_ch1 = pd.read_csv("/group/testa/Project/EndPoints/TPSSU/analysis/measurements_mix_2_ch1_NeuN.csv")

morphological_features = ['area', 'eccentricity', 'equivalent_diameter', 'extent',
                                            'major_axis_length', 'minor_axis_length', 'orientation', 'perimeter',
                                            'solidity']

tot_df_mix1_ch1['ImageName_ScanRegion_Rep'] = tot_df_mix1_ch1['ImageName'].apply(lambda x: '_'.join(x.split('_')[0:7]))

print(tot_df_mix1_ch1.shape)
tot_df_mix1_ch1 = tot_df_mix1_ch1.groupby('ImageName_ScanRegion_Rep').apply(filter_nuclei).reset_index(drop=True)
print(tot_df_mix1_ch1.shape)

tot_df_mix1_ch1['label_image'] = tot_df_mix1_ch1['label'].astype('str') + '_' + tot_df_mix1_ch1['ImageName_ScanRegion_Rep'].astype('str')
tot_df_mix1_ch1.set_index('label_image', inplace = True)

(4591259, 21)

/tmp/ipykernel_1121090/3325876105.py:11: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.
  tot_df_mix1_ch1 = tot_df_mix1_ch1.groupby('ImageName_ScanRegion_Rep').apply(filter_nuclei).reset_index(drop=True)

(4496846, 21)


tot_df_mix1_ch1 = tot_df_mix1_ch1[morphological_features]


tot_df_mix2_ch1['ImageName_ScanRegion_Rep'] = tot_df_mix2_ch1['ImageName'].apply(lambda x: '_'.join(x.split('_')[0:7]))

print(tot_df_mix2_ch1.shape)
tot_df_mix2_ch1 = tot_df_mix2_ch1.groupby('ImageName_ScanRegion_Rep').apply(filter_nuclei).reset_index(drop=True)
print(tot_df_mix2_ch1.shape)

tot_df_mix2_ch1['label_image'] = tot_df_mix2_ch1['label'].astype('str') + '_' + tot_df_mix2_ch1['ImageName_ScanRegion_Rep'].astype('str')
tot_df_mix2_ch1.set_index('label_image', inplace = True)

(4788573, 21)

/tmp/ipykernel_1121090/531573534.py:4: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.
  tot_df_mix2_ch1 = tot_df_mix2_ch1.groupby('ImageName_ScanRegion_Rep').apply(filter_nuclei).reset_index(drop=True)

(4690248, 21)


tot_df_mix2_ch1 = tot_df_mix2_ch1[morphological_features]


tot_df_morphological = pd.concat([tot_df_mix1_ch1, tot_df_mix2_ch1], axis=0)

tot_df_morphological['ImageName_ScanRegion_Rep'] = tot_df_morphological.reset_index()['label_image'].apply(lambda x: '_'.join(x.split('_')[1:8])).values
tot_df_morphological['ImageName_ScanRegion'] = tot_df_morphological.reset_index()['label_image'].apply(lambda x: '_'.join(x.split('_')[1:6])).values
tot_df_morphological['Condition_Line'] = tot_df_morphological['ImageName_ScanRegion'].apply(lambda x: sample_dict.get(x, 'Unknown'))
tot_df_morphological['Condition'] = tot_df_morphological['Condition_Line'].apply(lambda x: x.replace('_CTL04E', '').replace('_CTL08A', ''))
tot_df_morphological['Line'] = tot_df_morphological['Condition_Line'].apply(lambda x: x.split('_')[-1])

tot_df_morphological = tot_df_morphological[tot_df_morphological['Condition'].isin(hormonal_conditions)]


tot_df_morphological = tot_df_morphological[tot_df_morphological['ImageName_ScanRegion_Rep'].isin(good_tissue_to_keep_limit)]

tot_df_morphological = tot_df_morphological[tot_df_morphological["ImageName_ScanRegion_Rep"] != "20240424_manuel_lessi_0134_ScanRegion7_rep_1"] # outlier
tot_df_morphological = tot_df_morphological[tot_df_morphological['Condition'] != 'Unknown']  # filter out unknown conditions


features = tot_df_morphological[morphological_features]


len(tot_df_morphological["ImageName_ScanRegion_Rep"].unique())

139


# Standardize the features
scaler = StandardScaler()
features_scaled = scaler.fit_transform(features)

# Perform PCA
pca = PCA()  
principal_components = pca.fit_transform(features_scaled)


# Add the principal components to the dataframe
for i in range(principal_components.shape[1]):
    tot_df_morphological[f'PC{i+1}'] = principal_components[:, i]


# Extract the most important features for PCA1 and PCA2
pca_components = pd.DataFrame(pca.components_, columns=features.columns, index=[f'PC{i+1}' for i in range(len(pca.components_))])
important_features_pca1 = pca_components.loc['PC1'].abs().sort_values(ascending=False).head(5)
important_features_pca2 = pca_components.loc['PC2'].abs().sort_values(ascending=False).head(5)
important_features_pca3 = pca_components.loc['PC3'].abs().sort_values(ascending=False).head(5)

print("Top 5 features contributing to PCA1:")
print(important_features_pca1)

print("\nTop 5 features contributing to PCA2:")
print(important_features_pca2)

print("\nTop 5 features contributing to PCA3:")
print(important_features_pca3)

Top 5 features contributing to PCA1:
equivalent_diameter    0.451671
perimeter              0.446028
area                   0.445225
major_axis_length      0.430577
minor_axis_length      0.424077
Name: PC1, dtype: float64

Top 5 features contributing to PCA2:
extent               0.637839
solidity             0.531573
eccentricity         0.499679
major_axis_length    0.179739
minor_axis_length    0.132653
Name: PC2, dtype: float64

Top 5 features contributing to PCA3:
orientation          0.999614
solidity             0.023722
eccentricity         0.010693
extent               0.009049
major_axis_length    0.002369
Name: PC3, dtype: float64


grouped_df = tot_df_morphological.groupby('ImageName_ScanRegion_Rep').mean(['PC1', 'PC2']).reset_index()


grouped_df['ImageName_ScanRegion'] = grouped_df['ImageName_ScanRegion_Rep'].apply(lambda x: '_'.join(x.split('_')[0:5])).values
grouped_df['Condition_Line'] = grouped_df['ImageName_ScanRegion'].apply(lambda x: sample_dict.get(x, 'Unknown'))
grouped_df['Condition'] = grouped_df['Condition_Line'].apply(lambda x: x.replace('_CTL04E', '').replace('_CTL08A', ''))
grouped_df['Line'] = grouped_df['Condition_Line'].apply(lambda x: x.split('_')[-1])


cmap_conditions = plt.matplotlib.colors.ListedColormap([color_palette[c] for c in grouped_df['Condition'].astype('category').cat.categories])

plt.scatter(
    grouped_df['PC1'], grouped_df['PC2'],
    #alpha=0.3,
    c=grouped_df['Condition'].astype('category').cat.codes,
    cmap=cmap_conditions
)

plt.savefig('../../../../endpoints_figures/endpoints_PCA.pdf', dpi = 300)


plt.scatter(
    grouped_df['PC2'], grouped_df['PC3'],
    #alpha=0.3,
    c=grouped_df['Condition'].astype('category').cat.codes,
    cmap=cmap_conditions,
)
plt.savefig('../../../../endpoints_figures/endpoints_PCA_PC2_PC3.pdf', dpi = 300)


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['PC1']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['PC1'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['PC1']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['PC1'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('PC1')
ax[1].set_ylabel('PC1')

plt.savefig('../../../../endpoints_figures/endpoints_PC1_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_PC1_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['PC2']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['PC2'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['PC2']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['PC2'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('PC2')
ax[1].set_ylabel('PC2')

plt.savefig('../../../../endpoints_figures/endpoints_PC2_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_PC2_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['PC3']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['PC3'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['PC3']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['PC3'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('PC3')
ax[1].set_ylabel('PC3')

plt.savefig('../../../../endpoints_figures/endpoints_PC3_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_PC3_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['PC1']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['PC1'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL08A')],
    y='PC1', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['PC1']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['PC1'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[grouped_df['Line'] == 'CTL04E'],
    y='PC1', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('PC1')
ax[1].set_ylabel('PC1')

plt.savefig('../../../../endpoints_figures/endpoints_PC1_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_PC1_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']

to_test = {}
for i in grouped_df_CTL08A.Condition.unique():
    data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == i]['PC1'].values
    to_test[i] = data    
    
results = stats.kruskal(*list(to_test.values()))
results

KruskalResult(statistic=51.17330953153743, pvalue=3.87841286463078e-06)


DMSO = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == 'DMSO']['PC1'].values
results_MW_CTL08A_p_values = {}
statistics_MW_CTL08A = {}
for cond in grouped_df_CTL08A.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == cond]['PC1'].values
        results_MW_CTL08A_p_values[cond] = stats.mannwhitneyu(data, DMSO).pvalue
        statistics_MW_CTL08A[cond] = stats.mannwhitneyu(data, DMSO).statistic
        
pvals_CTL08A = list(results_MW_CTL08A_p_values.values())  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]
corrected_pvals_CTL08A = {i:j for i, j in zip(results_MW_CTL08A_p_values.keys(), corrected_pvals_CTL08A)}

results_p_val = pd.DataFrame({'CTL08A': corrected_pvals_CTL08A.values()}, index = corrected_pvals_CTL08A.keys())
results_stat = pd.DataFrame({'CTL08A': statistics_MW_CTL08A.values()}, index = statistics_MW_CTL08A.keys())


DMSO = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == 'DMSO']['PC1'].values
diff_with_DMSO = {}
for cond in grouped_df_CTL08A.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == cond]['PC1'].values
        diff_with_DMSO[cond] = np.median(data) - np.median(DMSO)


results_diff = pd.DataFrame({'CTL08A': diff_with_DMSO.values()}, index = diff_with_DMSO.keys())


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']

to_test = {}
for i in grouped_df_CTL04E.Condition.unique():
    data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == i]['PC1'].values
    to_test[i] = data    
    
results = stats.kruskal(*list(to_test.values()))
results

KruskalResult(statistic=42.534796404019005, pvalue=2.7062988752687628e-05)


DMSO = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == 'DMSO']['PC1'].values
results_MW_CTL04E_p_values = {}
statistics_MW_CTL04E = {}
for cond in grouped_df_CTL04E.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == cond]['PC1'].values
        results_MW_CTL04E_p_values[cond] = stats.mannwhitneyu(data, DMSO).pvalue
        statistics_MW_CTL04E[cond] = stats.mannwhitneyu(data, DMSO).statistic
        
pvals_CTL04E = list(results_MW_CTL04E_p_values.values())  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]
corrected_pvals_CTL04E = {i:j for i, j in zip(results_MW_CTL04E_p_values.keys(), corrected_pvals_CTL04E)}

results_p_val['CTL04E'] = results_p_val.index.map(corrected_pvals_CTL04E)
results_stat['CTL04E'] = results_stat.index.map(statistics_MW_CTL04E)


DMSO = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == 'DMSO']['PC1'].values
diff_with_DMSO = {}
for cond in grouped_df_CTL04E.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == cond]['PC1'].values
        diff_with_DMSO[cond] = np.median(data) - np.median(DMSO)


results_diff['CTL04E'] = pd.DataFrame({'CTL04E': diff_with_DMSO.values()}, index = diff_with_DMSO.keys())


results_diff = np.round(results_diff, 2)


fig, ax = plt.subplots(figsize = (4,7))
#results_stat = results_stat.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
#results_diff = results_diff[results_p_val < 0.05].fillna('')
sns.heatmap(results_p_val, annot=results_diff[results_p_val < 0.05].fillna(''), cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax=ax, fmt='')
plt.title('Condition vs DMSO\n')

Text(0.5, 1.0, 'Condition vs DMSO\n')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['PC2']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['PC2'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL08A')],
    y='PC2', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['PC2']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['PC2'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL04E')],
    y='PC2', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('PC2')
ax[1].set_ylabel('PC2')

plt.savefig('../../../../endpoints_figures/endpoints_PC2_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_PC2_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']

to_test = {}
for i in grouped_df_CTL08A.Condition.unique():
    data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == i]['PC2'].values
    to_test[i] = data    
    
results = stats.kruskal(*list(to_test.values()))
results

KruskalResult(statistic=57.05543654657589, pvalue=3.826203225448208e-07)


DMSO = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == 'DMSO']['PC2'].values
results_MW_CTL08A_p_values = {}
statistics_MW_CTL08A = {}
for cond in grouped_df_CTL08A.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == cond]['PC2'].values
        results_MW_CTL08A_p_values[cond] = stats.mannwhitneyu(data, DMSO).pvalue
        statistics_MW_CTL08A[cond] = stats.mannwhitneyu(data, DMSO).statistic
        
pvals_CTL08A = list(results_MW_CTL08A_p_values.values())  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]
corrected_pvals_CTL08A = {i:j for i, j in zip(results_MW_CTL08A_p_values.keys(), corrected_pvals_CTL08A)}

results_p_val = pd.DataFrame({'CTL08A': corrected_pvals_CTL08A.values()}, index = corrected_pvals_CTL08A.keys())
results_stat = pd.DataFrame({'CTL08A': statistics_MW_CTL08A.values()}, index = statistics_MW_CTL08A.keys())


DMSO = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == 'DMSO']['PC2'].values
diff_with_DMSO = {}
for cond in grouped_df_CTL08A.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == cond]['PC2'].values
        diff_with_DMSO[cond] = np.median(data) - np.median(DMSO)


results_diff = pd.DataFrame({'CTL08A': diff_with_DMSO.values()}, index = diff_with_DMSO.keys())


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']

to_test = {}
for i in grouped_df_CTL04E.Condition.unique():
    data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == i]['PC2'].values
    to_test[i] = data    
    
results = stats.kruskal(*list(to_test.values()))
results

KruskalResult(statistic=45.047435219460596, pvalue=1.0113944300737967e-05)


DMSO = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == 'DMSO']['PC2'].values
results_MW_CTL04E_p_values = {}
statistics_MW_CTL04E = {}
for cond in grouped_df_CTL04E.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == cond]['PC2'].values
        results_MW_CTL04E_p_values[cond] = stats.mannwhitneyu(data, DMSO).pvalue
        statistics_MW_CTL04E[cond] = stats.mannwhitneyu(data, DMSO).statistic
        
pvals_CTL04E = list(results_MW_CTL04E_p_values.values())  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]
corrected_pvals_CTL04E = {i:j for i, j in zip(results_MW_CTL04E_p_values.keys(), corrected_pvals_CTL04E)}

results_p_val['CTL04E'] = results_p_val.index.map(corrected_pvals_CTL04E)
results_stat['CTL04E'] = results_stat.index.map(statistics_MW_CTL04E)


DMSO = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == 'DMSO']['PC2'].values
diff_with_DMSO = {}
for cond in grouped_df_CTL04E.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == cond]['PC2'].values
        diff_with_DMSO[cond] = np.median(data) - np.median(DMSO)


results_diff['CTL04E'] = pd.DataFrame({'CTL04E': diff_with_DMSO.values()}, index = diff_with_DMSO.keys())


results_diff = np.round(results_diff, 2)


fig, ax = plt.subplots(figsize = (4,7))
#results_stat = results_stat.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
#results_diff = results_diff[results_p_val < 0.05].fillna('')
sns.heatmap(results_p_val, annot=results_diff[results_p_val < 0.05].fillna(''), cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax=ax, fmt='')
plt.title('Condition vs DMSO\n')

Text(0.5, 1.0, 'Condition vs DMSO\n')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['PC3']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['PC3'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL08A')],
    y='PC3', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['PC3']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['PC3'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL04E')],
    y='PC3', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('PC3')
ax[1].set_ylabel('PC3')

plt.savefig('../../../../endpoints_figures/endpoints_PC3_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_PC3_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']

to_test = {}
for i in grouped_df_CTL08A.Condition.unique():
    data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == i]['PC3'].values
    to_test[i] = data    
    
results = stats.kruskal(*list(to_test.values()))
results

KruskalResult(statistic=29.874099318403125, pvalue=0.00794236681106045)


DMSO = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == 'DMSO']['PC3'].values
results_MW_CTL08A_p_values = {}
statistics_MW_CTL08A = {}
for cond in grouped_df_CTL08A.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == cond]['PC3'].values
        results_MW_CTL08A_p_values[cond] = stats.mannwhitneyu(data, DMSO).pvalue
        statistics_MW_CTL08A[cond] = stats.mannwhitneyu(data, DMSO).statistic
        
pvals_CTL08A = list(results_MW_CTL08A_p_values.values())  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]
corrected_pvals_CTL08A = {i:j for i, j in zip(results_MW_CTL08A_p_values.keys(), corrected_pvals_CTL08A)}

results_p_val = pd.DataFrame({'CTL08A': corrected_pvals_CTL08A.values()}, index = corrected_pvals_CTL08A.keys())
results_stat = pd.DataFrame({'CTL08A': statistics_MW_CTL08A.values()}, index = statistics_MW_CTL08A.keys())


DMSO = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == 'DMSO']['PC3'].values
diff_with_DMSO = {}
for cond in grouped_df_CTL08A.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL08A[grouped_df_CTL08A['Condition'] == cond]['PC3'].values
        diff_with_DMSO[cond] = np.median(data) - np.median(DMSO)


results_diff = pd.DataFrame({'CTL08A': diff_with_DMSO.values()}, index = diff_with_DMSO.keys())


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']

to_test = {}
for i in grouped_df_CTL04E.Condition.unique():
    data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == i]['PC3'].values
    to_test[i] = data    
    
results = stats.kruskal(*list(to_test.values()))
results

KruskalResult(statistic=27.127445795875133, pvalue=0.007407220324972986)


DMSO = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == 'DMSO']['PC3'].values
results_MW_CTL04E_p_values = {}
statistics_MW_CTL04E = {}
for cond in grouped_df_CTL04E.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == cond]['PC3'].values
        results_MW_CTL04E_p_values[cond] = stats.mannwhitneyu(data, DMSO).pvalue
        statistics_MW_CTL04E[cond] = stats.mannwhitneyu(data, DMSO).statistic
        
pvals_CTL04E = list(results_MW_CTL04E_p_values.values())  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]
corrected_pvals_CTL04E = {i:j for i, j in zip(results_MW_CTL04E_p_values.keys(), corrected_pvals_CTL04E)}

results_p_val['CTL04E'] = results_p_val.index.map(corrected_pvals_CTL04E)
results_stat['CTL04E'] = results_stat.index.map(statistics_MW_CTL04E)


DMSO = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == 'DMSO']['PC3'].values
diff_with_DMSO = {}
for cond in grouped_df_CTL04E.Condition.unique():
    if cond != 'DMSO':
        data = grouped_df_CTL04E[grouped_df_CTL04E['Condition'] == cond]['PC3'].values
        diff_with_DMSO[cond] = np.median(data) - np.median(DMSO)


results_diff['CTL04E'] = pd.DataFrame({'CTL04E': diff_with_DMSO.values()}, index = diff_with_DMSO.keys())


results_diff = np.round(results_diff, 2)


fig, ax = plt.subplots(figsize = (4,7))
#results_stat = results_stat.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
#results_diff = results_diff[results_p_val < 0.05].fillna('')
sns.heatmap(results_p_val, annot=results_diff[results_p_val < 0.05].fillna(''), cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax=ax, fmt='')
plt.title('Condition vs DMSO\n')

Text(0.5, 1.0, 'Condition vs DMSO\n')


important_features_pca1 = pca_components.loc['PC1'].sort_values(ascending=False)
import matplotlib.pyplot as plt

# Lollipop plot for important_features_pca1
plt.figure(figsize=(5, 6))
x = important_features_pca1.index
y = important_features_pca1.values
plt.stem(x, y, basefmt=" ")
plt.xticks(rotation=90)
plt.ylabel('Contribution to PC1')
plt.tight_layout()
plt.savefig('../../../../endpoints_figures/important_features_pca1.pdf', dpi=300, bbox_inches='tight')


important_features_pca2 = pca_components.loc['PC2'].sort_values(ascending=False)

plt.figure(figsize=(5, 6))
x = important_features_pca2.index
y = important_features_pca2.values
plt.stem(x, y, basefmt=" ")
plt.xticks(rotation=90)
plt.ylabel('Contribution to PC2')
plt.tight_layout()
plt.savefig('../../../../endpoints_figures/important_features_pca2.pdf', dpi=300, bbox_inches='tight')


important_features_pca3 = pca_components.loc['PC3'].sort_values(ascending=False)


plt.figure(figsize=(5, 6))
x = important_features_pca3.index
y = important_features_pca3.values
plt.stem(x, y, basefmt=" ")
plt.xticks(rotation=90)
plt.ylabel('Contribution to PC3')
plt.tight_layout()
plt.savefig('../../../../endpoints_figures/important_features_pca3.pdf', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['equivalent_diameter']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['equivalent_diameter'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['equivalent_diameter']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['equivalent_diameter'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('equivalent_diameter')
ax[1].set_ylabel('equivalent_diameter')

plt.savefig('../../../../endpoints_figures/endpoints_equivalent_diameter_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_equivalent_diameter_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['equivalent_diameter']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['equivalent_diameter'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL08A')],
    y='equivalent_diameter', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['equivalent_diameter']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['equivalent_diameter'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL04E')],
    y='equivalent_diameter', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('equivalent_diameter')
ax[1].set_ylabel('equivalent_diameter')

plt.savefig('../../../../endpoints_figures/endpoints_equivalent_diameter_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_equivalent_diameter_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']
grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')

model_CTL08A = smf.ols("equivalent_diameter ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL08A,
                ).fit()

print(model_CTL08A.summary())

                             OLS Regression Results                            
===============================================================================
Dep. Variable:     equivalent_diameter   R-squared:                       0.541
Model:                             OLS   Adj. R-squared:                  0.439
Method:                  Least Squares   F-statistic:                     5.298
Date:                 Tue, 28 Apr 2026   Prob (F-statistic):           1.69e-06
Time:                         15:53:39   Log-Likelihood:                -193.55
No. Observations:                   78   AIC:                             417.1
Df Residuals:                       63   BIC:                             452.5
Df Model:                           14                                         
Covariance Type:             nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                19.1982      1.314     14.605      0.000      16.571      21.825
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]      1.3612      1.859      0.732      0.467      -2.354       5.076
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]     4.2478      2.277      1.866      0.067      -0.302       8.797
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]      2.4017      1.697      1.415      0.162      -0.989       5.793
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]     2.8381      1.859      1.527      0.132      -0.877       6.553
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]      2.8453      2.078      1.369      0.176      -1.308       6.999
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]     2.0426      2.078      0.983      0.329      -2.111       6.196
C(Condition, Treatment(reference='DMSO'))[T.GC_AG]        1.4172      2.078      0.682      0.498      -2.736       5.570
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]       2.6384      1.950      1.353      0.181      -1.258       6.535
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]        2.2393      1.859      1.205      0.233      -1.476       5.954
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]       0.8054      2.277      0.354      0.725      -3.744       5.355
C(Condition, Treatment(reference='DMSO'))[T.RET_AG]      12.6892      1.859      6.826      0.000       8.974      16.404
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]      0.2288      1.859      0.123      0.902      -3.486       3.944
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]      0.3718      1.950      0.191      0.849      -3.524       4.268
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]     0.3973      1.950      0.204      0.839      -3.499       4.293
==============================================================================
Omnibus:                       43.608   Durbin-Watson:                   1.275
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              633.742
Skew:                           1.052   Prob(JB):                    2.42e-138
Kurtosis:                      16.805   Cond. No.                         15.0
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/2352630096.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']
grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')

model_CTL04E = smf.ols("equivalent_diameter ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL04E,
                ).fit()

print(model_CTL04E.summary())

                             OLS Regression Results                            
===============================================================================
Dep. Variable:     equivalent_diameter   R-squared:                       0.738
Model:                             OLS   Adj. R-squared:                  0.672
Method:                  Least Squares   F-statistic:                     11.25
Date:                 Tue, 28 Apr 2026   Prob (F-statistic):           3.14e-10
Time:                         15:53:39   Log-Likelihood:                -73.088
No. Observations:                   61   AIC:                             172.2
Df Residuals:                       48   BIC:                             199.6
Df Model:                           12                                         
Covariance Type:             nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                20.3398      0.452     45.000      0.000      19.431      21.249
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]      0.8830      0.639      1.381      0.174      -0.402       2.168
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]     0.8536      0.690      1.236      0.222      -0.535       2.242
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]     -1.4807      0.584     -2.537      0.014      -2.654      -0.307
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]     1.6345      0.584      2.801      0.007       0.461       2.808
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]     -0.0834      0.690     -0.121      0.904      -1.472       1.305
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]    -1.1550      0.584     -1.979      0.054      -2.328       0.018
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]      -1.9279      0.584     -3.304      0.002      -3.101      -0.755
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]       -2.4619      0.606     -4.060      0.000      -3.681      -1.243
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]      -1.8504      0.584     -3.171      0.003      -3.024      -0.677
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]     -1.3852      0.639     -2.167      0.035      -2.670      -0.100
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]     -0.8752      0.584     -1.500      0.140      -2.048       0.298
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]     2.2845      0.783      2.918      0.005       0.710       3.859
==============================================================================
Omnibus:                        4.570   Durbin-Watson:                   2.574
Prob(Omnibus):                  0.102   Jarque-Bera (JB):                2.841
Skew:                          -0.334   Prob(JB):                        0.242
Kurtosis:                       2.180   Cond. No.                         15.1
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/2302572593.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')


resid_CTL08A = model_CTL08A.resid

stats.probplot(resid_CTL08A, dist="norm", plot=plt)
plt.show()


resid_CTL04E = model_CTL04E.resid

stats.probplot(resid_CTL04E, dist="norm", plot=plt)
plt.show()


pvals_CTL08A = model_CTL08A.pvalues[1:]  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]

df_results_CTL08A = pd.concat([model_CTL08A.params, pd.Series(corrected_pvals_CTL08A, index=model_CTL08A.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL08A.columns = ['intercept', 'pvalue']


pvals_CTL04E = model_CTL04E.pvalues[1:]  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]

df_results_CTL04E = pd.concat([model_CTL04E.params, pd.Series(corrected_pvals_CTL04E, index=model_CTL04E.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL04E.columns = ['intercept', 'pvalue']


pvals = pd.concat([df_results_CTL08A['pvalue'], df_results_CTL04E['pvalue']], axis = 1)
pvals.columns = ['CTL08A', 'CTL04E']
pvals.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in pvals.index]
pvals


intercepts = pd.concat([df_results_CTL08A['intercept'], df_results_CTL04E['intercept']], axis = 1)
intercepts.columns = ['CTL08A', 'CTL04E']
intercepts.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in intercepts.index]

params = model_CTL08A.params
mean_DMSO = params["Intercept"]
        
intercepts['CTL08A'] = intercepts['CTL08A']/ mean_DMSO * 100

params = model_CTL04E.params
mean_DMSO = params["Intercept"]
intercepts['CTL04E'] = intercepts['CTL04E']/ mean_DMSO * 100
intercepts


fig, ax = plt.subplots(figsize = (4,7))
intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
intercepts = intercepts[pvals < 0.05].fillna('')
sns.heatmap(pvals, annot=intercepts, fmt='', cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax = ax)
plt.title('Condition vs DMSO\n')
plt.savefig('../../../../endpoints_figures/endpoints_organoids_eq_diameter_OLS_test.png', dpi=300, bbox_inches='tight')

/tmp/ipykernel_1121090/3118655519.py:2: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['extent']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['extent'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['extent']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['extent'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('extent')
ax[1].set_ylabel('extent')

plt.savefig('../../../../endpoints_figures/endpoints_extent_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_extent_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['extent']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['extent'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

    sns.stripplot(
        data=grouped_df[(grouped_df['Line'] == 'CTL08A') & (grouped_df['Condition'] == cond)],
        y='extent', x='Condition', order=[cond], color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
    )

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['extent']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['extent'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

    sns.stripplot(
        data=grouped_df[(grouped_df['Line'] == 'CTL04E') & (grouped_df['Condition'] == cond)],
        y='extent', x='Condition', order=[cond], color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
    )

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('extent')
ax[1].set_ylabel('extent')

plt.savefig('../../../../endpoints_figures/endpoints_extent_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_extent_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']
grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')

model_CTL08A = smf.ols("extent ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL08A,
                ).fit()

print(model_CTL08A.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:                 extent   R-squared:                       0.483
Model:                            OLS   Adj. R-squared:                  0.368
Method:                 Least Squares   F-statistic:                     4.206
Date:                Tue, 28 Apr 2026   Prob (F-statistic):           3.73e-05
Time:                        15:54:29   Log-Likelihood:                 300.72
No. Observations:                  78   AIC:                            -571.4
Df Residuals:                      63   BIC:                            -536.1
Df Model:                          14                                         
Covariance Type:            nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                 0.7392      0.002    317.709      0.000       0.735       0.744
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]     -0.0058      0.003     -1.757      0.084      -0.012       0.001
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]    -0.0102      0.004     -2.528      0.014      -0.018      -0.002
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]     -0.0088      0.003     -2.938      0.005      -0.015      -0.003
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]    -0.0081      0.003     -2.464      0.016      -0.015      -0.002
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]     -0.0137      0.004     -3.719      0.000      -0.021      -0.006
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]     0.0004      0.004      0.099      0.922      -0.007       0.008
C(Condition, Treatment(reference='DMSO'))[T.GC_AG]        0.0008      0.004      0.219      0.828      -0.007       0.008
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]      -0.0054      0.003     -1.577      0.120      -0.012       0.001
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]       -0.0040      0.003     -1.211      0.230      -0.011       0.003
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]      -0.0083      0.004     -2.051      0.044      -0.016      -0.000
C(Condition, Treatment(reference='DMSO'))[T.RET_AG]      -0.0114      0.003     -3.479      0.001      -0.018      -0.005
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]      0.0041      0.003      1.234      0.222      -0.003       0.011
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]     -0.0069      0.003     -1.987      0.051      -0.014    3.79e-05
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]     0.0003      0.003      0.080      0.936      -0.007       0.007
==============================================================================
Omnibus:                       20.325   Durbin-Watson:                   2.053
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              146.730
Skew:                          -0.019   Prob(JB):                     1.37e-32
Kurtosis:                       9.719   Cond. No.                         15.0
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/3097047810.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']
grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')

model_CTL04E = smf.ols("extent ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL04E,
                ).fit()

print(model_CTL04E.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:                 extent   R-squared:                       0.696
Model:                            OLS   Adj. R-squared:                  0.620
Method:                 Least Squares   F-statistic:                     9.153
Date:                Tue, 28 Apr 2026   Prob (F-statistic):           8.30e-09
Time:                        15:54:29   Log-Likelihood:                 265.75
No. Observations:                  61   AIC:                            -505.5
Df Residuals:                      48   BIC:                            -478.1
Df Model:                          12                                         
Covariance Type:            nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                 0.7404      0.002    423.362      0.000       0.737       0.744
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]     -0.0087      0.002     -3.509      0.001      -0.014      -0.004
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]    -0.0073      0.003     -2.728      0.009      -0.013      -0.002
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]     -0.0007      0.002     -0.289      0.774      -0.005       0.004
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]    -0.0039      0.002     -1.739      0.088      -0.008       0.001
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]      0.0028      0.003      1.055      0.297      -0.003       0.008
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]     0.0085      0.002      3.756      0.000       0.004       0.013
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]       0.0031      0.002      1.366      0.178      -0.001       0.008
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]        0.0039      0.002      1.683      0.099      -0.001       0.009
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]       0.0020      0.002      0.893      0.376      -0.003       0.007
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]      0.0063      0.002      2.536      0.015       0.001       0.011
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]     -0.0025      0.002     -1.107      0.274      -0.007       0.002
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]     0.0006      0.003      0.197      0.844      -0.005       0.007
==============================================================================
Omnibus:                        0.005   Durbin-Watson:                   1.897
Prob(Omnibus):                  0.997   Jarque-Bera (JB):                0.138
Skew:                           0.010   Prob(JB):                        0.933
Kurtosis:                       2.768   Cond. No.                         15.1
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/270499386.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')


resid_CTL08A = model_CTL08A.resid

stats.probplot(resid_CTL08A, dist="norm", plot=plt)
plt.show()


resid_CTL04E = model_CTL04E.resid

stats.probplot(resid_CTL04E, dist="norm", plot=plt)
plt.show()


pvals_CTL08A = model_CTL08A.pvalues[1:]  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]

df_results_CTL08A = pd.concat([model_CTL08A.params, pd.Series(corrected_pvals_CTL08A, index=model_CTL08A.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL08A.columns = ['intercept', 'pvalue']


pvals_CTL04E = model_CTL04E.pvalues[1:]  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]

df_results_CTL04E = pd.concat([model_CTL04E.params, pd.Series(corrected_pvals_CTL04E, index=model_CTL04E.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL04E.columns = ['intercept', 'pvalue']


pvals = pd.concat([df_results_CTL08A['pvalue'], df_results_CTL04E['pvalue']], axis = 1)
pvals.columns = ['CTL08A', 'CTL04E']
pvals.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in pvals.index]
pvals


intercepts = pd.concat([df_results_CTL08A['intercept'], df_results_CTL04E['intercept']], axis = 1)
intercepts.columns = ['CTL08A', 'CTL04E']
intercepts.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in intercepts.index]

params = model_CTL08A.params
mean_DMSO = params["Intercept"]
        
intercepts['CTL08A'] = intercepts['CTL08A']/ mean_DMSO * 100

params = model_CTL04E.params
mean_DMSO = params["Intercept"]
intercepts['CTL04E'] = intercepts['CTL04E']/ mean_DMSO * 100
intercepts


fig, ax = plt.subplots(figsize = (4,7))
intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
intercepts = intercepts[pvals < 0.05].fillna('')
sns.heatmap(pvals, annot=intercepts, fmt='', cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax = ax)
plt.title('Condition vs DMSO\n')
plt.savefig('../../../../endpoints_figures/endpoints_organoids_extent_OLS_test.png', dpi=300, bbox_inches='tight')

/tmp/ipykernel_1121090/1658638270.py:2: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['eccentricity']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['eccentricity'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['eccentricity']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['eccentricity'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('eccentricity')
ax[1].set_ylabel('eccentricity')

plt.savefig('../../../../endpoints_figures/endpoints_eccentricity_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_eccentricity_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['eccentricity']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['eccentricity'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

    sns.stripplot(
        data=grouped_df[(grouped_df['Line'] == 'CTL08A') & (grouped_df['Condition'] == cond)],
        y='eccentricity', x='Condition', order=[cond], color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
    )

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['eccentricity']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['eccentricity'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

    sns.stripplot(
        data=grouped_df[(grouped_df['Line'] == 'CTL04E') & (grouped_df['Condition'] == cond)],
        y='eccentricity', x='Condition', order=[cond], color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
    )

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('eccentricity')
ax[1].set_ylabel('eccentricity')

plt.savefig('../../../../endpoints_figures/endpoints_eccentricity_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_eccentricity_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']
grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')

model_CTL08A = smf.ols("eccentricity ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL08A,
                ).fit()

print(model_CTL08A.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:           eccentricity   R-squared:                       0.622
Model:                            OLS   Adj. R-squared:                  0.537
Method:                 Least Squares   F-statistic:                     7.391
Date:                Tue, 28 Apr 2026   Prob (F-statistic):           8.35e-09
Time:                        15:55:15   Log-Likelihood:                 212.88
No. Observations:                  78   AIC:                            -395.8
Df Residuals:                      63   BIC:                            -360.4
Df Model:                          14                                         
Covariance Type:            nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                 0.6177      0.007     86.093      0.000       0.603       0.632
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]      0.0265      0.010      2.613      0.011       0.006       0.047
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]     0.0500      0.012      4.025      0.000       0.025       0.075
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]      0.0298      0.009      3.219      0.002       0.011       0.048
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]     0.0440      0.010      4.340      0.000       0.024       0.064
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]      0.0534      0.011      4.707      0.000       0.031       0.076
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]     0.0066      0.011      0.582      0.563      -0.016       0.029
C(Condition, Treatment(reference='DMSO'))[T.GC_AG]       -0.0062      0.011     -0.550      0.584      -0.029       0.016
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]       0.0116      0.011      1.086      0.281      -0.010       0.033
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]        0.0247      0.010      2.438      0.018       0.004       0.045
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]       0.0138      0.012      1.113      0.270      -0.011       0.039
C(Condition, Treatment(reference='DMSO'))[T.RET_AG]       0.0479      0.010      4.716      0.000       0.028       0.068
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]     -0.0066      0.010     -0.655      0.515      -0.027       0.014
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]      0.0148      0.011      1.394      0.168      -0.006       0.036
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]    -0.0106      0.011     -0.998      0.322      -0.032       0.011
==============================================================================
Omnibus:                       11.233   Durbin-Watson:                   1.655
Prob(Omnibus):                  0.004   Jarque-Bera (JB):               20.877
Skew:                          -0.449   Prob(JB):                     2.93e-05
Kurtosis:                       5.370   Cond. No.                         15.0
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/1012140979.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']
grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')

model_CTL04E = smf.ols("eccentricity ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL04E,
                ).fit()

print(model_CTL04E.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:           eccentricity   R-squared:                       0.699
Model:                            OLS   Adj. R-squared:                  0.624
Method:                 Least Squares   F-statistic:                     9.303
Date:                Tue, 28 Apr 2026   Prob (F-statistic):           6.47e-09
Time:                        15:55:15   Log-Likelihood:                 183.87
No. Observations:                  61   AIC:                            -341.7
Df Residuals:                      48   BIC:                            -314.3
Df Model:                          12                                         
Covariance Type:            nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                 0.6134      0.007     91.640      0.000       0.600       0.627
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]      0.0365      0.009      3.855      0.000       0.017       0.056
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]     0.0279      0.010      2.726      0.009       0.007       0.048
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]      0.0062      0.009      0.722      0.474      -0.011       0.024
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]     0.0093      0.009      1.072      0.289      -0.008       0.027
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]     -0.0094      0.010     -0.922      0.361      -0.030       0.011
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]    -0.0289      0.009     -3.346      0.002      -0.046      -0.012
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]      -0.0231      0.009     -2.668      0.010      -0.040      -0.006
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]       -0.0151      0.009     -1.678      0.100      -0.033       0.003
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]      -0.0109      0.009     -1.261      0.213      -0.028       0.006
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]     -0.0214      0.009     -2.266      0.028      -0.040      -0.002
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]      0.0056      0.009      0.652      0.517      -0.012       0.023
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]    -0.0029      0.012     -0.251      0.803      -0.026       0.020
==============================================================================
Omnibus:                        0.652   Durbin-Watson:                   2.177
Prob(Omnibus):                  0.722   Jarque-Bera (JB):                0.168
Skew:                          -0.022   Prob(JB):                        0.920
Kurtosis:                       3.253   Cond. No.                         15.1
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/2137971027.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')


resid_CTL08A = model_CTL08A.resid

stats.probplot(resid_CTL08A, dist="norm", plot=plt)
plt.show()


resid_CTL04E = model_CTL04E.resid

stats.probplot(resid_CTL04E, dist="norm", plot=plt)
plt.show()


pvals_CTL08A = model_CTL08A.pvalues[1:]  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]

df_results_CTL08A = pd.concat([model_CTL08A.params, pd.Series(corrected_pvals_CTL08A, index=model_CTL08A.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL08A.columns = ['intercept', 'pvalue']


pvals_CTL04E = model_CTL04E.pvalues[1:]  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]

df_results_CTL04E = pd.concat([model_CTL04E.params, pd.Series(corrected_pvals_CTL04E, index=model_CTL04E.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL04E.columns = ['intercept', 'pvalue']


pvals = pd.concat([df_results_CTL08A['pvalue'], df_results_CTL04E['pvalue']], axis = 1)
pvals.columns = ['CTL08A', 'CTL04E']
pvals.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in pvals.index]
pvals


intercepts = pd.concat([df_results_CTL08A['intercept'], df_results_CTL04E['intercept']], axis = 1)
intercepts.columns = ['CTL08A', 'CTL04E']
intercepts.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in intercepts.index]

params = model_CTL08A.params
mean_DMSO = params["Intercept"]
        
intercepts['CTL08A'] = intercepts['CTL08A']/ mean_DMSO * 100

params = model_CTL04E.params
mean_DMSO = params["Intercept"]
intercepts['CTL04E'] = intercepts['CTL04E']/ mean_DMSO * 100
intercepts


fig, ax = plt.subplots(figsize = (4,7))
intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
intercepts = intercepts[pvals < 0.05].fillna('')
sns.heatmap(pvals, annot=intercepts, fmt='', cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax = ax)
plt.title('Condition vs DMSO\n')
plt.savefig('../../../../endpoints_figures/endpoints_organoids_eccentricity_OLS_test.png', dpi=300, bbox_inches='tight')

/tmp/ipykernel_1121090/3100479939.py:2: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['orientation']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['orientation'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['orientation']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['orientation'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('orientation')
ax[1].set_ylabel('orientation')

plt.savefig('../../../../endpoints_figures/endpoints_orientation_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_orientation_all_nuclei.png', dpi=300, bbox_inches='tight')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['orientation']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['orientation'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

    sns.stripplot(
        data=grouped_df[(grouped_df['Line'] == 'CTL08A') & (grouped_df['Condition'] == cond)],
        y='orientation', x='Condition', order=[cond], color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
    )

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['orientation']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['orientation'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

    sns.stripplot(
        data=grouped_df[(grouped_df['Line'] == 'CTL04E') & (grouped_df['Condition'] == cond)],
        y='orientation', x='Condition', order=[cond], color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
    )

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('orientation')
ax[1].set_ylabel('orientation')

plt.savefig('../../../../endpoints_figures/endpoints_orientation_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_orientation_grouped.png', dpi=300, bbox_inches='tight')


tot_df_morphological['area_um2'] = tot_df_morphological['area'] * (0.325 ** 2)  # Convert area to um^2


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL08A') & 
                                 (tot_df_morphological['Condition'] == cond)]['area_um2']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL08A')]['area_um2'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

for i, cond in enumerate(hormonal_conditions):
    data = tot_df_morphological[(tot_df_morphological['Line'] == 'CTL04E') & 
                                 (tot_df_morphological['Condition'] == cond)]['area_um2']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = tot_df_morphological[(tot_df_morphological['Condition'] == 'DMSO') & 
                                       (tot_df_morphological['Line'] == 'CTL04E')]['area_um2'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('area_um2')
ax[1].set_ylabel('area_um2')

plt.savefig('../../../../endpoints_figures/endpoints_area_um2_all_nuclei.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_area_um2_all_nuclei.png', dpi=300, bbox_inches='tight')


grouped_df['area_um2'] = grouped_df['area'] * (0.325 ** 2)  # Convert area to um^2


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL08A') & 
                                 (grouped_df['Condition'] == cond)]['area_um2']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL08A')]['area_um2'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL08A')],
    y='area_um2', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = grouped_df[(grouped_df['Line'] == 'CTL04E') & 
                                 (grouped_df['Condition'] == cond)]['area_um2']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = grouped_df[(grouped_df['Condition'] == 'DMSO') & 
                                       (grouped_df['Line'] == 'CTL04E')]['area_um2'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=grouped_df[(grouped_df['Line'] == 'CTL04E')],
    y='area_um2', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('Nuclei area (um2)')
ax[1].set_ylabel('Nuclei area (um2)')

plt.savefig('../../../../endpoints_figures/endpoints_area_um2_grouped.pdf', dpi=300, bbox_inches='tight')
plt.savefig('../../../../endpoints_figures/endpoints_area_um2_grouped.png', dpi=300, bbox_inches='tight')


grouped_df_CTL08A = grouped_df[grouped_df['Line'] == 'CTL08A']
grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')
grouped_df_CTL08A["log_area"] = np.log(grouped_df_CTL08A["area_um2"])
grouped_df_CTL08A['log_area'].hist(bins = 10)

model_CTL08A = smf.ols("log_area ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL08A,
                ).fit()

print(model_CTL08A.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:               log_area   R-squared:                       0.590
Model:                            OLS   Adj. R-squared:                  0.499
Method:                 Least Squares   F-statistic:                     6.484
Date:                Tue, 28 Apr 2026   Prob (F-statistic):           7.59e-08
Time:                        15:56:45   Log-Likelihood:                 19.126
No. Observations:                  78   AIC:                            -8.251
Df Residuals:                      63   BIC:                             27.10
Df Model:                          14                                         
Covariance Type:            nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                 3.4752      0.086     40.402      0.000       3.303       3.647
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]      0.1418      0.122      1.165      0.248      -0.101       0.385
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]     0.4025      0.149      2.701      0.009       0.105       0.700
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]      0.2294      0.111      2.065      0.043       0.007       0.451
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]     0.2732      0.122      2.246      0.028       0.030       0.516
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]      0.2892      0.136      2.127      0.037       0.017       0.561
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]     0.2084      0.136      1.532      0.130      -0.063       0.480
C(Condition, Treatment(reference='DMSO'))[T.GC_AG]        0.1330      0.136      0.978      0.332      -0.139       0.405
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]       0.2446      0.128      1.917      0.060      -0.010       0.500
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]        0.2182      0.122      1.794      0.078      -0.025       0.461
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]       0.0831      0.149      0.558      0.579      -0.215       0.381
C(Condition, Treatment(reference='DMSO'))[T.RET_AG]       0.9205      0.122      7.567      0.000       0.677       1.164
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]      0.0280      0.122      0.230      0.819      -0.215       0.271
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]      0.0510      0.128      0.400      0.690      -0.204       0.306
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]     0.0367      0.128      0.288      0.774      -0.218       0.292
==============================================================================
Omnibus:                       29.224   Durbin-Watson:                   1.338
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              292.528
Skew:                           0.536   Prob(JB):                     3.01e-64
Kurtosis:                      12.427   Cond. No.                         15.0
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/2562141296.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL08A['Condition'] = grouped_df_CTL08A['Condition'].astype('category')
/tmp/ipykernel_1121090/2562141296.py:3: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL08A["log_area"] = np.log(grouped_df_CTL08A["area_um2"])


grouped_df_CTL04E = grouped_df[grouped_df['Line'] == 'CTL04E']
grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')
grouped_df_CTL04E["log_area"] = np.log(grouped_df_CTL04E["area_um2"])
grouped_df_CTL04E['log_area'].hist(bins = 10)

model_CTL04E = smf.ols("log_area ~ C(Condition, Treatment(reference='DMSO'))",
                data=grouped_df_CTL04E,
                ).fit()

print(model_CTL04E.summary())

                            OLS Regression Results                            
==============================================================================
Dep. Variable:               log_area   R-squared:                       0.709
Model:                            OLS   Adj. R-squared:                  0.636
Method:                 Least Squares   F-statistic:                     9.730
Date:                Tue, 28 Apr 2026   Prob (F-statistic):           3.22e-09
Time:                        15:56:45   Log-Likelihood:                 67.975
No. Observations:                  61   AIC:                            -110.0
Df Residuals:                      48   BIC:                            -82.51
Df Model:                          12                                         
Covariance Type:            nonrobust                                         
=========================================================================================================================
                                                            coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------------------------
Intercept                                                 3.5808      0.045     80.013      0.000       3.491       3.671
C(Condition, Treatment(reference='DMSO'))[T.ANDR_AG]      0.1083      0.063      1.712      0.093      -0.019       0.236
C(Condition, Treatment(reference='DMSO'))[T.ANDR_INH]     0.1091      0.068      1.596      0.117      -0.028       0.247
C(Condition, Treatment(reference='DMSO'))[T.ARYL_AG]     -0.1309      0.058     -2.266      0.028      -0.247      -0.015
C(Condition, Treatment(reference='DMSO'))[T.ARYL_INH]     0.1562      0.058      2.703      0.009       0.040       0.272
C(Condition, Treatment(reference='DMSO'))[T.ESTR_AG]      0.0240      0.068      0.350      0.728      -0.113       0.161
C(Condition, Treatment(reference='DMSO'))[T.ESTR_INH]    -0.0737      0.058     -1.275      0.208      -0.190       0.042
C(Condition, Treatment(reference='DMSO'))[T.GC_INH]      -0.1863      0.058     -3.224      0.002      -0.302      -0.070
C(Condition, Treatment(reference='DMSO'))[T.LX_AG]       -0.1903      0.060     -3.169      0.003      -0.311      -0.070
C(Condition, Treatment(reference='DMSO'))[T.LX_INH]      -0.1523      0.058     -2.636      0.011      -0.268      -0.036
C(Condition, Treatment(reference='DMSO'))[T.RET_INH]     -0.1244      0.063     -1.966      0.055      -0.252       0.003
C(Condition, Treatment(reference='DMSO'))[T.THYR_AG]     -0.0581      0.058     -1.006      0.320      -0.174       0.058
C(Condition, Treatment(reference='DMSO'))[T.THYR_INH]     0.2152      0.078      2.776      0.008       0.059       0.371
==============================================================================
Omnibus:                        5.053   Durbin-Watson:                   2.590
Prob(Omnibus):                  0.080   Jarque-Bera (JB):                3.099
Skew:                          -0.359   Prob(JB):                        0.212
Kurtosis:                       2.161   Cond. No.                         15.1
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

/tmp/ipykernel_1121090/4157921110.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL04E['Condition'] = grouped_df_CTL04E['Condition'].astype('category')
/tmp/ipykernel_1121090/4157921110.py:3: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  grouped_df_CTL04E["log_area"] = np.log(grouped_df_CTL04E["area_um2"])


resid_CTL08A = model_CTL08A.resid

stats.probplot(resid_CTL08A, dist="norm", plot=plt)
plt.show()


resid_CTL04E = model_CTL04E.resid

stats.probplot(resid_CTL04E, dist="norm", plot=plt)
plt.show()


pvals_CTL08A = model_CTL08A.pvalues[1:]  # exclude intercept
corrected_pvals_CTL08A = multipletests(pvals_CTL08A, method='fdr_bh')[1]

df_results_CTL08A = pd.concat([model_CTL08A.params, pd.Series(corrected_pvals_CTL08A, index=model_CTL08A.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL08A.columns = ['intercept', 'pvalue']


pvals_CTL04E = model_CTL04E.pvalues[1:]  # exclude intercept
corrected_pvals_CTL04E = multipletests(pvals_CTL04E, method='fdr_bh')[1]

df_results_CTL04E = pd.concat([model_CTL04E.params, pd.Series(corrected_pvals_CTL04E, index=model_CTL04E.params.index[1:])], axis = 1).iloc[1:]
df_results_CTL04E.columns = ['intercept', 'pvalue']


pvals = pd.concat([df_results_CTL08A['pvalue'], df_results_CTL04E['pvalue']], axis = 1)
pvals.columns = ['CTL08A', 'CTL04E']
pvals.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in pvals.index]
pvals


intercepts = pd.concat([df_results_CTL08A['intercept'], df_results_CTL04E['intercept']], axis = 1)
intercepts.columns = ['CTL08A', 'CTL04E']
intercepts.index = [i.split('C(Condition, Treatment(reference=\'DMSO\'))[T.')[1].replace(']', '') for i in intercepts.index]
intercepts = (np.exp(intercepts) - 1) * 100
intercepts


fig, ax = plt.subplots(figsize = (4,7))
intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")
intercepts = intercepts[pvals < 0.05].fillna('')
sns.heatmap(pvals, annot=intercepts, fmt='', cmap='Greens_r', cbar_kws={'label': 'p-value'}, vmin=0, vmax=0.05, ax = ax)
plt.title('Condition vs DMSO\n')
plt.savefig('../../../../endpoints_figures/endpoints_organoids_nuclei_area_OLS_test.png', dpi=300, bbox_inches='tight')

/tmp/ipykernel_1121090/2973917183.py:2: FutureWarning: DataFrame.applymap has been deprecated. Use DataFrame.map instead.
  intercepts = intercepts.applymap(lambda x: f"{x:+.1f}%" if pd.notnull(x) else "")


organoid_areas = pd.read_csv('/group/testa/Project/EndPoints/TPSSU/analysis/organoids_area.csv', index_col=0)
organoid_areas


organoid_areas['ImageName_ScanRegion'] = organoid_areas.reset_index()['index'].apply(lambda x: '_'.join(x.split('_')[0:5])).values
organoid_areas['Condition_Line'] = organoid_areas['ImageName_ScanRegion'].map(sample_dict)
organoid_areas['Line'] = organoid_areas['Condition_Line'].apply(lambda x: x.split('_')[-1] if pd.notnull(x) else x)
organoid_areas['Condition'] = organoid_areas['Condition_Line'].str.replace('_CTL08A', '', regex=True)
organoid_areas['Condition'] = organoid_areas['Condition'].str.replace('_CTL04E', '', regex=True)
organoid_areas = organoid_areas[organoid_areas['Condition'].isin(hormonal_conditions)]
organoid_areas


organoid_areas['area_um2'] = organoid_areas['area'] * (0.325 ** 2)  # Convert area to um^2

/tmp/ipykernel_1121090/172599718.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['area_um2'] = organoid_areas['area'] * (0.325 ** 2)  # Convert area to um^2


summed_area = tot_df_morphological[['ImageName_ScanRegion_Rep', 'area_um2']].groupby('ImageName_ScanRegion_Rep').sum()


organoid_areas['tot_nuclei_areas'] = summed_area

/tmp/ipykernel_1121090/1334222756.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['tot_nuclei_areas'] = summed_area


organoid_areas['nuclear_area'] = organoid_areas['tot_nuclei_areas'] / organoid_areas['area_um2'] * 100

/tmp/ipykernel_1121090/3125954352.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['nuclear_area'] = organoid_areas['tot_nuclei_areas'] / organoid_areas['area_um2'] * 100


organoid_areas


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = organoid_areas[(organoid_areas['Line'] == 'CTL08A') & 
                                 (organoid_areas['Condition'] == cond)]['nuclear_area']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = organoid_areas[(organoid_areas['Condition'] == 'DMSO') & 
                                       (organoid_areas['Line'] == 'CTL08A')]['nuclear_area'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_areas[(organoid_areas['Line'] == 'CTL08A')],
    y='nuclear_area', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = organoid_areas[(organoid_areas['Line'] == 'CTL04E') & 
                                 (organoid_areas['Condition'] == cond)]['nuclear_area']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = organoid_areas[(organoid_areas['Condition'] == 'DMSO') & 
                                       (organoid_areas['Line'] == 'CTL04E')]['nuclear_area'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_areas[(organoid_areas['Line'] == 'CTL04E')],
    y='nuclear_area', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('nuclear_area')
ax[1].set_ylabel('nuclear_area')

Text(0, 0.5, 'nuclear_area')


measurements_files = glob.glob("/group/testa/Project/EndPoints/TPSSU/analysis/measurements/*.csv")
measurements_files_mix_1 = [f for f in measurements_files if any(name in f for name in mix_1_names)]


tot_df_mix1_ch2 = pd.DataFrame()
measurements_files_mix_1_ch2 = [f for f in measurements_files_mix_1 if 'channel_2' in f]

for file in measurements_files_mix_1_ch2:
    print(file)
    df = pd.read_csv(file, index_col=0)
    df['ImageName'] = file.split("/")[-1].split(".")[0]
    print(df.shape)
    tot_df_mix1_ch2 = pd.concat([tot_df_mix1_ch2, df], axis=0)

/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion3_rep_1_channel_2_measures.csv
(14032, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion2_rep_0_channel_2_measures.csv
(36257, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion3_rep_1_channel_2_measures.csv
(29165, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion2_rep_0_channel_2_measures.csv
(28234, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion1_rep_2_channel_2_measures.csv
(5186, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion2_rep_1_channel_2_measures.csv
(30598, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion0_rep_0_channel_2_measures.csv
(4420, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion4_rep_0_channel_2_measures.csv
(5637, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion5_rep_2_channel_2_measures.csv
(44878, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion2_rep_1_channel_2_measures.csv
(25453, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion5_rep_0_channel_2_measures.csv
(14547, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion2_rep_0_channel_2_measures.csv
(65200, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion4_rep_1_channel_2_measures.csv
(18423, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion0_rep_2_channel_2_measures.csv
(12038, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion0_rep_1_channel_2_measures.csv
(52451, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion0_rep_0_channel_2_measures.csv
(69681, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion3_rep_0_channel_2_measures.csv
(60277, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion7_rep_0_channel_2_measures.csv
(36423, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion2_rep_1_channel_2_measures.csv
(54724, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion3_rep_2_channel_2_measures.csv
(48396, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion1_rep_0_channel_2_measures.csv
(18479, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion5_rep_0_channel_2_measures.csv
(61473, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion3_rep_2_channel_2_measures.csv
(62393, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion5_rep_2_channel_2_measures.csv
(27998, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion4_rep_0_channel_2_measures.csv
(16418, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion0_rep_0_channel_2_measures.csv
(54097, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion0_rep_1_channel_2_measures.csv
(19128, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion2_rep_2_channel_2_measures.csv
(7446, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion0_rep_0_channel_2_measures.csv
(6943, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion3_rep_2_channel_2_measures.csv
(44660, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion2_rep_0_channel_2_measures.csv
(6134, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion2_rep_0_channel_2_measures.csv
(43655, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion5_rep_0_channel_2_measures.csv
(55824, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion4_rep_0_channel_2_measures.csv
(43161, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion4_rep_2_channel_2_measures.csv
(26186, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion5_rep_2_channel_2_measures.csv
(42766, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion1_rep_0_channel_2_measures.csv
(32479, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion0_rep_2_channel_2_measures.csv
(46684, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion1_rep_1_channel_2_measures.csv
(7889, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion7_rep_1_channel_2_measures.csv
(1, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion3_rep_1_channel_2_measures.csv
(4878, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion5_rep_1_channel_2_measures.csv
(24682, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion3_rep_1_channel_2_measures.csv
(31993, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion3_rep_2_channel_2_measures.csv
(48939, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion4_rep_2_channel_2_measures.csv
(1405, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion0_rep_2_channel_2_measures.csv
(50003, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion4_rep_1_channel_2_measures.csv
(26358, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion1_rep_1_channel_2_measures.csv
(4685, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion0_rep_1_channel_2_measures.csv
(42864, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion4_rep_1_channel_2_measures.csv
(12551, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion1_rep_0_channel_2_measures.csv
(128250, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion2_rep_2_channel_2_measures.csv
(68825, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion0_rep_1_channel_2_measures.csv
(10615, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion2_rep_2_channel_2_measures.csv
(54944, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion2_rep_1_channel_2_measures.csv
(32605, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion2_rep_0_channel_2_measures.csv
(23699, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion3_rep_1_channel_2_measures.csv
(20919, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion4_rep_2_channel_2_measures.csv
(33055, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion3_rep_0_channel_2_measures.csv
(70503, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion1_rep_0_channel_2_measures.csv
(12349, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion1_rep_1_channel_2_measures.csv
(30782, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion6_rep_0_channel_2_measures.csv
(22787, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion4_rep_0_channel_2_measures.csv
(42538, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion5_rep_1_channel_2_measures.csv
(31370, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion3_rep_0_channel_2_measures.csv
(4465, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion0_rep_1_channel_2_measures.csv
(24746, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion1_rep_1_channel_2_measures.csv
(5061, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion5_rep_2_channel_2_measures.csv
(45127, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion1_rep_1_channel_2_measures.csv
(12207, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion5_rep_0_channel_2_measures.csv
(36668, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion4_rep_0_channel_2_measures.csv
(22478, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion0_rep_0_channel_2_measures.csv
(75915, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion4_rep_0_channel_2_measures.csv
(881, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion1_rep_2_channel_2_measures.csv
(25173, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion0_rep_1_channel_2_measures.csv
(9836, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion7_rep_2_channel_2_measures.csv
(27770, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion5_rep_0_channel_2_measures.csv
(52956, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion2_rep_2_channel_2_measures.csv
(55563, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion1_rep_1_channel_2_measures.csv
(92340, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion2_rep_1_channel_2_measures.csv
(2185, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion2_rep_2_channel_2_measures.csv
(13307, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion3_rep_0_channel_2_measures.csv
(47582, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion0_rep_2_channel_2_measures.csv
(1122, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion1_rep_0_channel_2_measures.csv
(5646, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion6_rep_1_channel_2_measures.csv
(22272, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion5_rep_1_channel_2_measures.csv
(59825, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion3_rep_0_channel_2_measures.csv
(11258, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion4_rep_2_channel_2_measures.csv
(6712, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion1_rep_0_channel_2_measures.csv
(26007, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion7_rep_3_channel_2_measures.csv
(72679, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion2_rep_1_channel_2_measures.csv
(17407, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion6_rep_2_channel_2_measures.csv
(4246, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion5_rep_1_channel_2_measures.csv
(79266, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion4_rep_1_channel_2_measures.csv
(10935, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion2_rep_0_channel_2_measures.csv
(24890, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion0_rep_1_channel_2_measures.csv
(24601, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion5_rep_2_channel_2_measures.csv
(58504, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion2_rep_2_channel_2_measures.csv
(7832, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion2_rep_1_channel_2_measures.csv
(194410, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion1_rep_0_channel_2_measures.csv
(2649, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion0_rep_2_channel_2_measures.csv
(35739, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion5_rep_1_channel_2_measures.csv
(72165, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion4_rep_0_channel_2_measures.csv
(79123, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion4_rep_1_channel_2_measures.csv
(17120, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion3_rep_0_channel_2_measures.csv
(43389, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion3_rep_2_channel_2_measures.csv
(19504, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0131_ScanRegion4_rep_1_channel_2_measures.csv
(38346, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0128_ScanRegion0_rep_0_channel_2_measures.csv
(70479, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0134_ScanRegion3_rep_0_channel_2_measures.csv
(50144, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion3_rep_2_channel_2_measures.csv
(1195, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion3_rep_1_channel_2_measures.csv
(54110, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0133_ScanRegion4_rep_1_channel_2_measures.csv
(4832, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0132_ScanRegion3_rep_1_channel_2_measures.csv
(39112, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0130_ScanRegion0_rep_0_channel_2_measures.csv
(41176, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion3_rep_3_channel_2_measures.csv
(9818, 19)
/group/testa/Project/EndPoints/TPSSU/analysis/measurements/20240424_manuel_lessi_0129_ScanRegion3_rep_4_channel_2_measures.csv
(23465, 19)


tot_pos_SOX2 = {}
density_SOX2 = {}
n_nuclei = {}
n_nuclei_tot_pos_SOX2 = {}
tot_area_SOX2 = {}
tot_area_no_SOX2 = {}

for file in measurements_files_mix_1_ch2:
    #print(file)
    df = pd.read_csv(file, index_col=0)
    
    imageName_scanRegion_rep = file.split('/')[-1].split('.')[0].replace('_channel_2_measures', '')
    imageName_scanRegion = '_'.join(imageName_scanRegion_rep.split('_')[:5])
    
    if imageName_scanRegion_rep in good_tissue_to_keep_limit and imageName_scanRegion in sample_sheet['ImageName_ScanRegion'].values:
        
        df['ImageName'] = file.split("/")[-1].split(".")[0]
        tot_pos_SOX2[file.split("/")[-1].split(".")[0]] = {}
        density_SOX2[imageName_scanRegion_rep] = {}
        n_nuclei_tot_pos_SOX2[imageName_scanRegion_rep] = {}

        df['area_um2'] = df['area'] * (0.325 ** 2)  # Convert area to um^2

        min_quantile = np.quantile(df['area_um2'], 0.01)

        max_quantile = np.quantile(df['area_um2'], 0.99)

        df = df[(df['area_um2'] > min_quantile) & (df['area_um2'] < max_quantile)]

        n_nuclei[file.split("/")[-1].split(".")[0]] = df.shape[0]

        otsu_thrs = filters.threshold_multiotsu(np.array(df['intensity_mean']), classes = 4)

        for i, thr in enumerate(otsu_thrs):
            tot_pos_SOX2[file.split("/")[-1].split(".")[0]][f'otsu_{i}'] = np.sum(df['intensity_mean'] > thr) / df.shape[0] * 100
            n_nuclei_tot_pos_SOX2[imageName_scanRegion_rep][f'otsu_{i}'] = np.sum(df['intensity_mean'] > thr)

            if i == 1:
                tot_area_SOX2[imageName_scanRegion_rep] = df[df['intensity_mean'] > thr]['area_um2'].sum()
                tot_area_no_SOX2[imageName_scanRegion_rep] = df[df['intensity_mean'] <= thr]['area_um2'].sum()


organoid_areas['SOX2_pos_area'] = organoid_areas.index.map(tot_area_SOX2)
organoid_areas['SOX2_neg_area'] = organoid_areas.index.map(tot_area_no_SOX2)

/tmp/ipykernel_1121090/4008132345.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['SOX2_pos_area'] = organoid_areas.index.map(tot_area_SOX2)
/tmp/ipykernel_1121090/4008132345.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['SOX2_neg_area'] = organoid_areas.index.map(tot_area_no_SOX2)


organoid_areas['SOX2_pos_area_over_tot'] = organoid_areas['SOX2_pos_area'] / organoid_areas['area_um2'] * 100
organoid_areas['SOX2_neg_area_over_tot'] = organoid_areas['SOX2_neg_area'] / organoid_areas['area_um2'] * 100

/tmp/ipykernel_1121090/1300664090.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['SOX2_pos_area_over_tot'] = organoid_areas['SOX2_pos_area'] / organoid_areas['area_um2'] * 100
/tmp/ipykernel_1121090/1300664090.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  organoid_areas['SOX2_neg_area_over_tot'] = organoid_areas['SOX2_neg_area'] / organoid_areas['area_um2'] * 100


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

organoid_area_mix1 = organoid_areas[~organoid_areas['SOX2_pos_area'].isna()]

for i, cond in enumerate(hormonal_conditions):
    data = organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL08A') & 
                                 (organoid_area_mix1['Condition'] == cond)]['nuclear_area']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = organoid_area_mix1[(organoid_area_mix1['Condition'] == 'DMSO') & 
                                       (organoid_area_mix1['Line'] == 'CTL08A')]['nuclear_area'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL08A')],
    y='nuclear_area', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL04E') & 
                                 (organoid_area_mix1['Condition'] == cond)]['nuclear_area']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = organoid_area_mix1[(organoid_area_mix1['Condition'] == 'DMSO') & 
                                       (organoid_area_mix1['Line'] == 'CTL04E')]['nuclear_area'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL04E')],
    y='nuclear_area', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('nuclear_area/organoid area')
ax[1].set_ylabel('nuclear_area/organoid area')

Text(0, 0.5, 'nuclear_area/organoid area')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

organoid_area_mix1 = organoid_areas[~organoid_areas['SOX2_pos_area'].isna()]

for i, cond in enumerate(hormonal_conditions):
    data = organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL08A') & 
                                 (organoid_area_mix1['Condition'] == cond)]['SOX2_pos_area_over_tot']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = organoid_area_mix1[(organoid_area_mix1['Condition'] == 'DMSO') & 
                                       (organoid_area_mix1['Line'] == 'CTL08A')]['SOX2_pos_area_over_tot'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL08A')],
    y='SOX2_pos_area_over_tot', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL04E') & 
                                 (organoid_area_mix1['Condition'] == cond)]['SOX2_pos_area_over_tot']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = organoid_area_mix1[(organoid_area_mix1['Condition'] == 'DMSO') & 
                                       (organoid_area_mix1['Line'] == 'CTL04E')]['SOX2_pos_area_over_tot'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL04E')],
    y='SOX2_pos_area_over_tot', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('SOX2_pos_area_over_tot')
ax[1].set_ylabel('SOX2_pos_area_over_tot')

Text(0, 0.5, 'SOX2_pos_area_over_tot')


fig, ax = plt.subplots(1, 2, figsize=(20, 5))

for i, cond in enumerate(hormonal_conditions):
    data = organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL08A') & 
                                 (organoid_area_mix1['Condition'] == cond)]['SOX2_neg_area_over_tot']
    box = ax[0].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
        # Add stripplot for CTL08A (single points per box)

    dmso_median = organoid_area_mix1[(organoid_area_mix1['Condition'] == 'DMSO') & 
                                       (organoid_area_mix1['Line'] == 'CTL08A')]['SOX2_neg_area_over_tot'].median()
    ax[0].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL08A')],
    y='SOX2_neg_area_over_tot', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[0], size=8
)

for i, cond in enumerate(hormonal_conditions):
    data = organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL04E') & 
                                 (organoid_area_mix1['Condition'] == cond)]['SOX2_neg_area_over_tot']
    box = ax[1].boxplot(data, positions=[i], patch_artist=True, widths=.8)
    for patch in box['boxes']:
        patch.set_facecolor(color_palette[cond])
        box['medians'][0].set_color('black')
    dmso_median = organoid_area_mix1[(organoid_area_mix1['Condition'] == 'DMSO') & 
                                       (organoid_area_mix1['Line'] == 'CTL04E')]['SOX2_neg_area_over_tot'].median()
    ax[1].axhline(dmso_median, color='red', linestyle='--', linewidth=.7, alpha=0.7)

sns.stripplot(
    data=organoid_area_mix1[(organoid_area_mix1['Line'] == 'CTL04E')],
    y='SOX2_neg_area_over_tot', x='Condition', order=hormonal_conditions, color='black', alpha=0.5, jitter=False, ax=ax[1], size=8
)

_ = ax[0].set_xticks(range(len(hormonal_conditions)))
_ = ax[0].set_xticklabels(hormonal_conditions, rotation=90)

_ = ax[1].set_xticks(range(len(hormonal_conditions)))
_ = ax[1].set_xticklabels(hormonal_conditions, rotation=90)

ax[0].set_title('CTL08A')
ax[1].set_title('CTL04E')

ax[0].set_ylabel('SOX2_neg_area_over_tot')
ax[1].set_ylabel('SOX2_neg_area_over_tot')

Text(0, 0.5, 'SOX2_neg_area_over_tot')

	Unnamed: 0	area	label	intensity_mean	intensity_min	intensity_max
0	20240424_manuel_lessi_0133_ScanRegion4_rep_0_c...	2415914.0	1.0	8478.139486	5323.0	45060.0
1	20240424_manuel_lessi_0143_ScanRegion0_rep_1_c...	10287703.0	1.0	9564.326972	6720.0	42038.0
2	20240424_manuel_lessi_0133_ScanRegion2_rep_1_c...	3015477.0	1.0	11510.699631	5536.0	49132.0
3	20240424_manuel_lessi_0128_ScanRegion4_rep_1_c...	6843311.0	1.0	6301.567833	4239.0	34600.0
4	20240424_manuel_lessi_0141_ScanRegion0_rep_2_c...	23305924.0	1.0	5729.083352	3988.0	24114.0
...	...	...	...	...	...	...
278	20240424_manuel_lessi_0145_ScanRegion2_rep_1_c...	35809244.0	1.0	367.353397	340.0	2983.0
279	20240424_manuel_lessi_0134_ScanRegion6_rep_1_c...	NaN	NaN	NaN	NaN	NaN
280	20240424_manuel_lessi_0144_ScanRegion2_rep_1_c...	6015146.0	1.0	7573.352427	4627.0	44177.0
281	20240424_manuel_lessi_0129_ScanRegion5_rep_1_c...	23404593.0	1.0	7714.037972	4597.0	65534.0
282	20240424_manuel_lessi_0145_ScanRegion0_rep_1_c...	10141827.0	1.0	388.299189	360.0	10460.0

	Unnamed: 0	area	label	intensity_mean	intensity_min	intensity_max	OriginalImageName	ImageName_ScanRegion	ImageName_ScanRegion_Rep
0	20240424_manuel_lessi_0133_ScanRegion4_rep_0_c...	2415914.0	1.0	8478.139486	5323.0	45060.0	20240424_manuel_lessi_0133	20240424_manuel_lessi_0133_ScanRegion4	20240424_manuel_lessi_0133_ScanRegion4_rep_0
1	20240424_manuel_lessi_0143_ScanRegion0_rep_1_c...	10287703.0	1.0	9564.326972	6720.0	42038.0	20240424_manuel_lessi_0143	20240424_manuel_lessi_0143_ScanRegion0	20240424_manuel_lessi_0143_ScanRegion0_rep_1
2	20240424_manuel_lessi_0133_ScanRegion2_rep_1_c...	3015477.0	1.0	11510.699631	5536.0	49132.0	20240424_manuel_lessi_0133	20240424_manuel_lessi_0133_ScanRegion2	20240424_manuel_lessi_0133_ScanRegion2_rep_1
3	20240424_manuel_lessi_0128_ScanRegion4_rep_1_c...	6843311.0	1.0	6301.567833	4239.0	34600.0	20240424_manuel_lessi_0128	20240424_manuel_lessi_0128_ScanRegion4	20240424_manuel_lessi_0128_ScanRegion4_rep_1
4	20240424_manuel_lessi_0141_ScanRegion0_rep_2_c...	23305924.0	1.0	5729.083352	3988.0	24114.0	20240424_manuel_lessi_0141	20240424_manuel_lessi_0141_ScanRegion0	20240424_manuel_lessi_0141_ScanRegion0_rep_2
...	...	...	...	...	...	...	...	...	...
278	20240424_manuel_lessi_0145_ScanRegion2_rep_1_c...	35809244.0	1.0	367.353397	340.0	2983.0	20240424_manuel_lessi_0145	20240424_manuel_lessi_0145_ScanRegion2	20240424_manuel_lessi_0145_ScanRegion2_rep_1
279	20240424_manuel_lessi_0134_ScanRegion6_rep_1_c...	NaN	NaN	NaN	NaN	NaN	20240424_manuel_lessi_0134	20240424_manuel_lessi_0134_ScanRegion6	20240424_manuel_lessi_0134_ScanRegion6_rep_1
280	20240424_manuel_lessi_0144_ScanRegion2_rep_1_c...	6015146.0	1.0	7573.352427	4627.0	44177.0	20240424_manuel_lessi_0144	20240424_manuel_lessi_0144_ScanRegion2	20240424_manuel_lessi_0144_ScanRegion2_rep_1
281	20240424_manuel_lessi_0129_ScanRegion5_rep_1_c...	23404593.0	1.0	7714.037972	4597.0	65534.0	20240424_manuel_lessi_0129	20240424_manuel_lessi_0129_ScanRegion5	20240424_manuel_lessi_0129_ScanRegion5_rep_1
282	20240424_manuel_lessi_0145_ScanRegion0_rep_1_c...	10141827.0	1.0	388.299189	360.0	10460.0	20240424_manuel_lessi_0145	20240424_manuel_lessi_0145_ScanRegion0	20240424_manuel_lessi_0145_ScanRegion0_rep_1

	Unnamed: 0	area	label	intensity_mean	intensity_min	intensity_max	OriginalImageName	ImageName_ScanRegion	ImageName_ScanRegion_Rep
0	20240424_manuel_lessi_0133_ScanRegion4_rep_0_c...	2415914.0	1.0	8478.139486	5323.0	45060.0	20240424_manuel_lessi_0133	20240424_manuel_lessi_0133_ScanRegion4	20240424_manuel_lessi_0133_ScanRegion4_rep_0
2	20240424_manuel_lessi_0133_ScanRegion2_rep_1_c...	3015477.0	1.0	11510.699631	5536.0	49132.0	20240424_manuel_lessi_0133	20240424_manuel_lessi_0133_ScanRegion2	20240424_manuel_lessi_0133_ScanRegion2_rep_1
3	20240424_manuel_lessi_0128_ScanRegion4_rep_1_c...	6843311.0	1.0	6301.567833	4239.0	34600.0	20240424_manuel_lessi_0128	20240424_manuel_lessi_0128_ScanRegion4	20240424_manuel_lessi_0128_ScanRegion4_rep_1
9	20240424_manuel_lessi_0134_ScanRegion6_rep_0_c...	7147296.0	1.0	5035.728892	2796.0	20250.0	20240424_manuel_lessi_0134	20240424_manuel_lessi_0134_ScanRegion6	20240424_manuel_lessi_0134_ScanRegion6_rep_0
12	20240424_manuel_lessi_0130_ScanRegion2_rep_1_c...	5736151.0	1.0	3811.331029	2366.0	59754.0	20240424_manuel_lessi_0130	20240424_manuel_lessi_0130_ScanRegion2	20240424_manuel_lessi_0130_ScanRegion2_rep_1
...	...	...	...	...	...	...	...	...	...
272	20240424_manuel_lessi_0131_ScanRegion2_rep_1_c...	619563.0	1.0	7816.673183	4931.0	56017.0	20240424_manuel_lessi_0131	20240424_manuel_lessi_0131_ScanRegion2	20240424_manuel_lessi_0131_ScanRegion2_rep_1
274	20240424_manuel_lessi_0129_ScanRegion5_rep_2_c...	15482008.0	1.0	5670.165540	3481.0	56907.0	20240424_manuel_lessi_0129	20240424_manuel_lessi_0129_ScanRegion5	20240424_manuel_lessi_0129_ScanRegion5_rep_2
277	20240424_manuel_lessi_0133_ScanRegion0_rep_2_c...	10075425.0	1.0	4029.633461	2629.0	41430.0	20240424_manuel_lessi_0133	20240424_manuel_lessi_0133_ScanRegion0	20240424_manuel_lessi_0133_ScanRegion0_rep_2
279	20240424_manuel_lessi_0134_ScanRegion6_rep_1_c...	NaN	NaN	NaN	NaN	NaN	20240424_manuel_lessi_0134	20240424_manuel_lessi_0134_ScanRegion6	20240424_manuel_lessi_0134_ScanRegion6_rep_1
281	20240424_manuel_lessi_0129_ScanRegion5_rep_1_c...	23404593.0	1.0	7714.037972	4597.0	65534.0	20240424_manuel_lessi_0129	20240424_manuel_lessi_0129_ScanRegion5	20240424_manuel_lessi_0129_ScanRegion5_rep_1

	Unnamed: 0	area	label	intensity_mean	intensity_min	intensity_max	OriginalImageName	ImageName_ScanRegion	ImageName_ScanRegion_Rep
1	20240424_manuel_lessi_0143_ScanRegion0_rep_1_c...	10287703.0	1.0	9564.326972	6720.0	42038.0	20240424_manuel_lessi_0143	20240424_manuel_lessi_0143_ScanRegion0	20240424_manuel_lessi_0143_ScanRegion0_rep_1
4	20240424_manuel_lessi_0141_ScanRegion0_rep_2_c...	23305924.0	1.0	5729.083352	3988.0	24114.0	20240424_manuel_lessi_0141	20240424_manuel_lessi_0141_ScanRegion0	20240424_manuel_lessi_0141_ScanRegion0_rep_2
5	20240424_manuel_lessi_0138_ScanRegion3_rep_1_c...	6139706.0	1.0	10728.699008	7679.0	47373.0	20240424_manuel_lessi_0138	20240424_manuel_lessi_0138_ScanRegion3	20240424_manuel_lessi_0138_ScanRegion3_rep_1
6	20240424_manuel_lessi_0140_ScanRegion2_rep_0_c...	2887437.0	1.0	10582.573429	8073.0	32277.0	20240424_manuel_lessi_0140	20240424_manuel_lessi_0140_ScanRegion2	20240424_manuel_lessi_0140_ScanRegion2_rep_0
10	20240424_manuel_lessi_0142_ScanRegion3_rep_0_c...	18501332.0	1.0	8751.319145	6006.0	40819.0	20240424_manuel_lessi_0142	20240424_manuel_lessi_0142_ScanRegion3	20240424_manuel_lessi_0142_ScanRegion3_rep_0
...	...	...	...	...	...	...	...	...	...
266	20240424_manuel_lessi_0139_ScanRegion0_rep_2_c...	16192847.0	1.0	12416.619954	8449.0	52637.0	20240424_manuel_lessi_0139	20240424_manuel_lessi_0139_ScanRegion0	20240424_manuel_lessi_0139_ScanRegion0_rep_2
268	20240424_manuel_lessi_0140_ScanRegion4_rep_1_c...	10125138.0	1.0	6630.255008	3940.0	59013.0	20240424_manuel_lessi_0140	20240424_manuel_lessi_0140_ScanRegion4	20240424_manuel_lessi_0140_ScanRegion4_rep_1
271	20240424_manuel_lessi_0137_ScanRegion4_rep_1_c...	10843902.0	1.0	6203.065238	4195.0	41839.0	20240424_manuel_lessi_0137	20240424_manuel_lessi_0137_ScanRegion4	20240424_manuel_lessi_0137_ScanRegion4_rep_1
273	20240424_manuel_lessi_0142_ScanRegion2_rep_1_c...	2596930.0	1.0	4422.110883	3233.0	18355.0	20240424_manuel_lessi_0142	20240424_manuel_lessi_0142_ScanRegion2	20240424_manuel_lessi_0142_ScanRegion2_rep_1
276	20240424_manuel_lessi_0141_ScanRegion4_rep_1_c...	26594153.0	1.0	6924.010593	4440.0	41733.0	20240424_manuel_lessi_0141	20240424_manuel_lessi_0141_ScanRegion4	20240424_manuel_lessi_0141_ScanRegion4_rep_1

	CTL08A	CTL04E
ANDR_AG	6.969339e-01	0.208263
ANDR_INH	4.218924e-01	0.242570
ARYL_AG	4.218924e-01	0.028941
ARYL_INH	4.218924e-01	0.017565
ESTR_AG	4.218924e-01	0.904346
ESTR_INH	5.765536e-01	0.080280
GC_AG	6.969339e-01	NaN
GC_INH	4.218924e-01	0.010585
LX_AG	4.657286e-01	0.002164
LX_INH	9.024162e-01	0.010585
RET_AG	5.618575e-08	NaN
RET_INH	9.024162e-01	0.060399
THYR_AG	9.024162e-01	0.186927
THYR_INH	9.024162e-01	0.016037

Load manual checks¶

Load sample sheet¶

Load measurements file names¶

Not filtered - only morphological features - PCA on single nuclei¶

Grouped by mean - PC1¶

Statistical test - PC1¶

CTL08A¶

CTL04E¶

Grouped by mean - PC2¶

Statistical test - PC2¶

CTL08A¶

CTL04E¶

Grouped by mean - PC3¶

Statistical test - PC3¶

CTL08A¶

CTL04E¶

Lollipop plots¶

Plot of equivalent diameter¶

Grouped by mean per organoid¶

Statistical test - Equivalent diameter¶

Plot of extent¶

Grouped by mean per organoid¶

Statistical test - Extent¶

Plot of eccentricity¶

Grouped by mean per organoid¶

Statistical test - Eccentricity¶

Plot of orientation¶

Grouped by mean per organoid¶

Plot of nuclei area¶

Grouped by mean per organoid¶

Statistical test - Area¶

Organoid areas¶

	Path	Scan Area	Condition	Line	OPA	Paraffin Block ID	MIX
0	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	ESTR_AG	CTL04E	12	29	1
1	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	ESTR_AG	CTL08A	12	30	1
2	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	ESTR_INH	CTL04E	12	31	1
3	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	ANDR_AG	CTL04E	12	33	1
4	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	ESTR_INH	CTL08A	12	32	1
5	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	ANDR_AG	CTL08A	12	34	1
6	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	ANDR_ANT	CTL04E	13	35	1
7	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	ARYL_INH	CTL08A	13	40	1
8	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	ARYL_INH	CTL04E	13	39	1
9	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	ARYL_AG	CTL08A	13	38	1
10	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	ANDR_ANT	CTL08A	13	36	1
11	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	ARYL_AG	CTL04E	13	37	1
12	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	LX_AG	CTL04E	14	41	1
13	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	GC_AG	CTL08A	14	46	1
14	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	LX_AG	CTL08A	14	42	1
15	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	LX_INH	CTL04E	14	43	1
16	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	LX_INH	CTL08A	14	44	1
17	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	GC_INH	CTL04E	15	47	1
18	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	THYR_INH	CTL04E	15	51	1
19	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	THYR_AG	CTL08A	15	50	1
20	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	GC_INH	CTL08A	15	48	1
21	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	THYR_AG	CTL04E	15	49	1
22	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	THYR_INH	CTL08A	15	52	1
23	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	DMSO	CTL04E	16	57	1
24	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	DMSO	CTL08A	16	58	1
25	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	RET_AG	CTL08A	17	64	1
26	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	RET_INH	CTL04E	18	65	1
27	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	RET_INH	CTL08A	18	66	1
28	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	ESTR_AG	CTL04E	12	29	2
29	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	ANDR_AG	CTL08A	12	34	2
30	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	ESTR_AG	CTL08A	12	30	2
31	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	ESTR_INH	CTL04E	12	31	2
32	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	ANDR_AG	CTL04E	12	33	2
33	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	ESTR_INH	CTL08A	12	32	2
34	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	ANDR_INH	CTL08A	13	36	2
35	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	ANDR_INH	CTL04E	13	35	2
36	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	ARYL_INH	CTL04E	13	39	2
37	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	ARYL_AG	CTL08A	13	38	2
38	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	ARYL_INH	CTL08A	13	40	2
39	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	ARYL_AG	CTL04E	13	37	2
40	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	LX_AG	CTL08A	14	42	2
41	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	LX_INH	CTL08A	14	44	2
42	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	LX_AG	CTL04E	14	41	2
43	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	LX_INH	CTL04E	14	43	2
44	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	GC_AG	CTL08A	14	46	2
45	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	GC_INH	CTL08A	15	48	2
46	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	GC_INH	CTL04E	15	47	2
47	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	THYR_INH	CTL04E	15	51	2
48	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	THYR_AG	CTL04E	15	49	2
49	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion4	THYR_INH	CTL08A	15	52	2
50	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion5	THYR_AG	CTL08A	15	50	2
51	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	DMSO	CTL08A	16	58	2
52	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion3	DMSO	CTL04E	16	57	2
53	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion2	RET_AG	CTL08A	17	64	2
54	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion0	RET_INH	CTL08A	18	66	2
55	T:\Project\EndPoints\TPSSU\Pictures\20240424_m...	ScanRegion1	RET_INH	CTL04E	18	65	2

	CTL08A	CTL04E
ANDR_AG	7.090013	4.341349
ANDR_INH	22.125814	4.196708
ARYL_AG	12.510013	-7.279584
ARYL_INH	14.783145	8.036174
ESTR_AG	14.820896	-0.410092
ESTR_INH	10.639776	-5.678729
GC_AG	7.382001	NaN
GC_INH	13.743015	-9.478257
LX_AG	11.664044	-12.103826
LX_INH	4.195281	-9.097445
RET_AG	66.096060	NaN
RET_INH	1.192013	-6.810068
THYR_AG	1.936849	-4.302990
THYR_INH	2.069517	11.231696

	CTL08A	CTL04E
ANDR_AG	0.146588	0.005935
ANDR_INH	0.046107	0.035509
ARYL_AG	0.021496	0.844374
ARYL_INH	0.046107	0.197766
ESTR_AG	0.005981	0.395654
ESTR_INH	0.936429	0.005601
GC_AG	0.936429	NaN
GC_INH	0.186370	0.305775
LX_AG	0.293340	0.197766
LX_INH	0.102465	0.451261
RET_AG	0.006429	NaN
RET_INH	0.293340	0.043522
THYR_AG	0.102465	0.395654
THYR_INH	0.936429	0.844461

	CTL08A	CTL04E
ANDR_AG	-0.782122	-1.172067
ANDR_INH	-1.378235	-0.984255
ARYL_AG	-1.193967	-0.088054
ARYL_INH	-1.096948	-0.530394
ESTR_AG	-1.851045	0.380626
ESTR_INH	0.049108	1.145418
GC_AG	0.108898	NaN
GC_INH	-0.736219	0.416484
LX_AG	-0.538977	0.533323
LX_INH	-1.117992	0.272461
RET_AG	-1.548628	NaN
RET_INH	0.549067	0.847269
THYR_AG	-0.927813	-0.337672
THYR_INH	0.037385	0.080701

	CTL08A	CTL04E
ANDR_AG	0.026175	0.004128
ANDR_INH	0.000545	0.031130
ARYL_AG	0.005701	0.564293
ARYL_INH	0.000246	0.433723
ESTR_AG	0.000100	0.481559
ESTR_INH	0.584482	0.009591
GC_AG	0.584482	NaN
GC_INH	0.393980	0.031130
LX_AG	0.035222	0.199756
LX_INH	0.393980	0.365859
RET_AG	0.000100	NaN
RET_INH	0.584482	0.067244
THYR_AG	0.294518	0.564293
THYR_INH	0.409853	0.802613

	CTL08A	CTL04E
ANDR_AG	4.291810	5.948758
ANDR_INH	8.096875	4.543271
ARYL_AG	4.826548	1.016616
ARYL_INH	7.128947	1.509976
ESTR_AG	8.644100	-1.536757
ESTR_INH	1.069257	-4.713942
GC_AG	-1.009511	NaN
GC_INH	1.871787	-3.758465
LX_AG	4.004447	-2.456410
LX_INH	2.238430	-1.776391
RET_AG	7.746693	NaN
RET_INH	-1.076113	-3.496482
THYR_AG	2.401156	0.919046
THYR_INH	-1.719633	-0.475073

	CTL08A	CTL04E
ANDR_AG	3.861730e-01	0.140144
ANDR_INH	6.201731e-02	0.155963
ARYL_AG	1.204013e-01	0.055995
ARYL_INH	1.204013e-01	0.027070
ESTR_AG	1.204013e-01	0.727503
ESTR_INH	2.282874e-01	0.250067
GC_AG	4.645990e-01	NaN
GC_INH	1.393849e-01	0.015985
LX_AG	1.553794e-01	0.015985
LX_INH	7.368958e-01	0.027070
RET_AG	2.860717e-09	NaN
RET_INH	8.188953e-01	0.094554
THYR_AG	8.055180e-01	0.348721
THYR_INH	8.188953e-01	0.027070

	CTL08A	CTL04E
ANDR_AG	15.230298	11.441127
ANDR_INH	49.550300	11.530624
ARYL_AG	25.779160	-12.271330
ARYL_INH	31.417438	16.903754
ESTR_AG	33.541879	2.424938
ESTR_INH	23.170375	-7.102512
GC_AG	14.224804	NaN
GC_INH	27.712485	-16.995242
LX_AG	24.379871	-17.325295
LX_INH	8.664554	-14.124428
RET_AG	151.044298	NaN
RET_INH	2.836367	-11.697077
THYR_AG	5.236901	-5.644258
THYR_INH	3.741964	24.006274

	area
20240424_manuel_lessi_0141_ScanRegion1_rep_0	48618374
20240424_manuel_lessi_0131_ScanRegion0_rep_2	36008220
20240424_manuel_lessi_0133_ScanRegion4_rep_0	8422522
20240424_manuel_lessi_0143_ScanRegion2_rep_1	61481135
20240424_manuel_lessi_0147_ScanRegion0_rep_1	162545379
...	...
20240424_manuel_lessi_0129_ScanRegion1_rep_0	26572184
20240424_manuel_lessi_0129_ScanRegion0_rep_0	9292198
20240424_manuel_lessi_0142_ScanRegion1_rep_2	38017446
20240424_manuel_lessi_0129_ScanRegion2_rep_1	29459387
20240424_manuel_lessi_0134_ScanRegion4_rep_1	22517768

	area	ImageName_ScanRegion	Condition_Line	Line	Condition
20240424_manuel_lessi_0131_ScanRegion0_rep_2	36008220	20240424_manuel_lessi_0131_ScanRegion0	GC_INH_CTL04E	CTL04E	GC_INH
20240424_manuel_lessi_0133_ScanRegion4_rep_0	8422522	20240424_manuel_lessi_0133_ScanRegion4	RET_AG_CTL08A	CTL08A	RET_AG
20240424_manuel_lessi_0140_ScanRegion4_rep_2	32680044	20240424_manuel_lessi_0140_ScanRegion4	THYR_INH_CTL08A	CTL08A	THYR_INH
20240424_manuel_lessi_0140_ScanRegion0_rep_1	36008064	20240424_manuel_lessi_0140_ScanRegion0	GC_INH_CTL08A	CTL08A	GC_INH
20240424_manuel_lessi_0138_ScanRegion3_rep_1	16957648	20240424_manuel_lessi_0138_ScanRegion3	ARYL_AG_CTL08A	CTL08A	ARYL_AG
...	...	...	...	...	...
20240424_manuel_lessi_0131_ScanRegion4_rep_1	35725975	20240424_manuel_lessi_0131_ScanRegion4	THYR_AG_CTL04E	CTL04E	THYR_AG
20240424_manuel_lessi_0140_ScanRegion3_rep_0	46296832	20240424_manuel_lessi_0140_ScanRegion3	THYR_AG_CTL04E	CTL04E	THYR_AG
20240424_manuel_lessi_0128_ScanRegion4_rep_0	42198648	20240424_manuel_lessi_0128_ScanRegion4	ESTR_INH_CTL08A	CTL08A	ESTR_INH
20240424_manuel_lessi_0129_ScanRegion1_rep_0	26572184	20240424_manuel_lessi_0129_ScanRegion1	ARYL_INH_CTL08A	CTL08A	ARYL_INH
20240424_manuel_lessi_0129_ScanRegion2_rep_1	29459387	20240424_manuel_lessi_0129_ScanRegion2	ARYL_INH_CTL04E	CTL04E	ARYL_INH