In [1]:
import pandas as pd
import os
from scipy.stats import hypergeom
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
from venn import venn
from itertools import combinations
from matplotlib.colors import LinearSegmentedColormap
from collections import OrderedDict
import pyreadr
In [2]:
os.chdir('')

GRNs¶

In [3]:
GRN_iPSC = pd.read_csv("GRN_hiPSC.csv")
GRN_iPSC_filt = GRN_iPSC.loc[(GRN_iPSC['coef_abs'] > 0.01) & (GRN_iPSC['-logp'] >= 2)]

GRN_iMeLC = pd.read_csv("GRN_iMeLC.csv")
GRN_iMeLC_filt = GRN_iMeLC.loc[(GRN_iMeLC['coef_abs'] > 0.01) & (GRN_iMeLC['-logp'] >= 2)]

GRN_hPGCLC = pd.read_csv("GRN_hPGCLC.csv")
GRN_PGCLC_filt = GRN_hPGCLC.loc[(GRN_hPGCLC['coef_abs'] > 0.01) & (GRN_hPGCLC['-logp'] >= 2)]

GRN_hEGCLC = pd.read_csv("GRN_hEGCLC.csv")
GRN_EGCLC_filt = GRN_hEGCLC.loc[(GRN_hEGCLC['coef_abs'] > 0.01) & (GRN_hEGCLC['-logp'] >= 2)]
In [4]:
GRN_iPSC_filt.to_csv('GRN_hiPSC_filt.csv', index=False)
GRN_iMeLC_filt.to_csv('GRN_iMeLC_filt.csv', index=False)
GRN_PGCLC_filt.to_csv('GRN_hPGCLC_filt.csv', index=False)
GRN_EGCLC_filt.to_csv('GRN_hEGCLC_filt.csv', index=False)
In [5]:
def dist_plot2(number_targets):
    quartiles = number_targets.quantile([0.25, 0.5, 0.75])
    q1, median, q3 = quartiles[0.25], quartiles[0.5], quartiles[0.75]

    plt.figure(figsize=(10, 6))
    plt.hist(number_targets, bins=10, edgecolor='black', color='skyblue')
    plt.axvline(q1, color='r', linestyle='dashed', linewidth=1, label=f'1st Quartile (Q1): {q1}')
    plt.axvline(median, color='g', linestyle='dashed', linewidth=1, label=f'Median (Q2): {median}')
    plt.axvline(q3, color='b', linestyle='dashed', linewidth=1, label=f'3st Quartile (Q3): {q3}')
    plt.xlabel('Number of Target per TF')
    plt.ylabel('Frequency')
    plt.legend()
    plt.grid(True)
    plt.show()
In [6]:
number_targets_IP = GRN_iPSC_filt.groupby('source')['target'].nunique()
number_targets_ME = GRN_iMeLC_filt.groupby('source')['target'].nunique()
number_targets_PG = GRN_PGCLC_filt.groupby('source')['target'].nunique()
number_targets_EG = GRN_EGCLC_filt.groupby('source')['target'].nunique()
In [7]:
dist_plot2(number_targets_IP)
dist_plot2(number_targets_ME)
dist_plot2(number_targets_PG)
dist_plot2(number_targets_EG)
In [8]:
tf_IP_10targets = number_targets_IP[number_targets_IP >= 10].index
GRN_iPSC_filt = GRN_iPSC_filt[GRN_iPSC_filt['source'].isin(tf_IP_10targets)]
GRN_iPSC_filt
Out[8]:
Unnamed: 0 source target coef_mean coef_abs p -logp
13 13 EGR1 A4GALT -0.011337 0.011337 2.955064e-11 10.529433
21 21 FOS A4GALT 0.011398 0.011398 1.493396e-09 8.825825
24 24 FOXP4 A4GALT 0.023915 0.023915 4.643730e-14 13.333133
25 25 SNAI1 A4GALT 0.010837 0.010837 2.070093e-13 12.684010
29 29 ELF3 A4GALT 0.013303 0.013303 5.879207e-13 12.230681
... ... ... ... ... ... ... ...
350434 350434 ZNF165 ZYG11A 0.013430 0.013430 9.960155e-12 11.001734
350451 350451 BHLHE40 ZYG11A 0.011117 0.011117 1.016146e-08 7.993044
350461 350461 ATF3 ZYG11A 0.014585 0.014585 5.512019e-15 14.258689
350462 350462 ZIC3 ZYG11A 0.022429 0.022429 7.037211e-19 18.152599
350472 350472 BCL3 ZYG11A 0.016571 0.016571 5.683222e-15 14.245405

77471 rows × 7 columns

In [9]:
tf_ME_10targets = number_targets_ME[number_targets_ME >= 10].index
GRN_iMeLC_filt = GRN_iMeLC_filt[GRN_iMeLC_filt['source'].isin(tf_ME_10targets)]
GRN_iMeLC_filt
Out[9]:
Unnamed: 0 source target coef_mean coef_abs p -logp
3 3 MAF A4GALT 0.010437 0.010437 9.697370e-13 12.013346
12 12 TFCP2 A4GALT 0.011424 0.011424 6.991379e-13 12.155437
45 45 POLR3G A4GALT 0.013872 0.013872 1.354756e-11 10.868139
50 50 ZFP57 A4GALT 0.016220 0.016220 2.714614e-15 14.566292
56 56 OVOL2 A4GALT 0.016652 0.016652 6.244700e-14 13.204488
... ... ... ... ... ... ... ...
350434 350434 ZNF165 ZYG11A 0.017363 0.017363 4.585605e-12 11.338603
350441 350441 FOXH1 ZYG11A 0.023062 0.023062 2.598020e-14 13.585357
350450 350450 REST ZYG11A 0.014005 0.014005 1.156650e-14 13.936798
350452 350452 ZIC2 ZYG11A -0.015993 0.015993 2.509253e-11 10.600456
350463 350463 TFAP2C ZYG11A 0.011303 0.011303 1.144868e-13 12.941244

61767 rows × 7 columns

In [10]:
tf_PG_10targets = number_targets_PG[number_targets_PG >= 10].index
GRN_PGCLC_filt = GRN_PGCLC_filt[GRN_PGCLC_filt['source'].isin(tf_PG_10targets)]
GRN_PGCLC_filt
Out[10]:
Unnamed: 0 source target coef_mean coef_abs p -logp
4 4 E2F1 A4GALT -0.011982 0.011982 1.966049e-09 8.706406
6 6 RFX3 A4GALT -0.012635 0.012635 7.385996e-09 8.131591
9 9 KLF4 A4GALT 0.027695 0.027695 1.407787e-13 12.851463
11 11 SP8 A4GALT -0.010359 0.010359 1.637664e-10 9.785775
14 14 FOXB2 A4GALT 0.010374 0.010374 1.412924e-10 9.849881
... ... ... ... ... ... ... ...
350450 350450 REST ZYG11A 0.016032 0.016032 3.348030e-18 17.475211
350453 350453 MEF2A ZYG11A 0.010208 0.010208 9.384473e-14 13.027590
350458 350458 NPAS1 ZYG11A 0.011459 0.011459 6.519771e-13 12.185768
350463 350463 TFAP2C ZYG11A 0.020360 0.020360 2.841033e-19 18.546524
350474 350474 ZNF816 ZYG11A 0.010861 0.010861 1.269887e-14 13.896235

28207 rows × 7 columns

In [11]:
tf_EG_10targets = number_targets_EG[number_targets_EG >= 10].index
GRN_EGCLC_filt = GRN_EGCLC_filt[GRN_EGCLC_filt['source'].isin(tf_EG_10targets)]
GRN_EGCLC_filt
Out[11]:
Unnamed: 0 source target coef_mean coef_abs p -logp
4 4 E2F1 A4GALT 0.012186 0.012186 2.743329e-10 9.561722
10 10 MYB A4GALT -0.011644 0.011644 1.174329e-06 5.930210
13 13 EGR1 A4GALT -0.021030 0.021030 2.029755e-09 8.692556
15 15 ZNF143 A4GALT -0.010884 0.010884 1.777862e-10 9.750102
33 33 KLF2 A4GALT -0.010645 0.010645 5.090827e-11 10.293212
... ... ... ... ... ... ... ...
350359 350359 NPAS1 ZSWIM7 -0.013866 0.013866 3.799716e-13 12.420249
350363 350363 PBX2 ZSWIM7 0.046887 0.046887 2.019494e-13 12.694757
350367 350367 KLF6 ZSWIM7 0.033844 0.033844 2.047703e-14 13.688733
350415 350415 TCF7L2 ZYG11A 0.010945 0.010945 1.534989e-10 9.813895
350450 350450 REST ZYG11A 0.011057 0.011057 4.454638e-18 17.351188

82575 rows × 7 columns

DMGs¶

In [12]:
DM_hPGCLC_iPSC = pd.read_excel('DMRAnnotated/DMRs_hPGCLCsvshiPSCs.xlsx')
DM_hPGCLC_hEGCLC = pd.read_excel('DMRAnnotated/DMRs_hPGCLCsvshEGCLCs.xlsx')
DM_iPSC_hEGCLC = pd.read_excel('DMRAnnotated/DMRs_hiPSCsvshEGCLCs.xlsx')
In [13]:
tot_geni_DM_PGvsiPSC = DM_hPGCLC_iPSC['hgnc_symbol'].unique().tolist()
tot_geni_DM_PGvsEG = DM_hPGCLC_hEGCLC['hgnc_symbol'].unique().tolist()
tot_geni_DM_iPSCvsEG = DM_iPSC_hEGCLC['hgnc_symbol'].unique().tolist()

k-means DEGs¶

In [14]:
cluster0_kmeans = pd.read_excel('genes_in_cluster_0.xlsx')
cluster1_kmeans = pd.read_excel('genes_in_cluster_1.xlsx')
cluster2_kmeans = pd.read_excel('genes_in_cluster_2.xlsx')
cluster3_kmeans = pd.read_excel('genes_in_cluster_3.xlsx')
cluster4_kmeans = pd.read_excel('genes_in_cluster_4.xlsx')
cluster5_kmeans = pd.read_excel('genes_in_cluster_5.xlsx')
cluster6_kmeans = pd.read_excel('genes_in_cluster_6.xlsx')
cluster7_kmeans = pd.read_excel('genes_in_cluster_7.xlsx')
cluster8_kmeans = pd.read_excel('genes_in_cluster_8.xlsx')
cluster9_kmeans = pd.read_excel('genes_in_cluster_9.xlsx')
cluster10_kmeans = pd.read_excel('genes_in_cluster_10.xlsx')
cluster11_kmeans = pd.read_excel('genes_in_cluster_11.xlsx')
cluster12_kmeans = pd.read_excel('genes_in_cluster_12.xlsx')
cluster13_kmeans = pd.read_excel('genes_in_cluster_13.xlsx')
cluster14_kmeans = pd.read_excel('genes_in_cluster_14.xlsx')
In [15]:
cluster0_gene = cluster0_kmeans['Unnamed: 0'].to_list()
cluster1_gene = cluster1_kmeans['Unnamed: 0'].to_list()
cluster2_gene = cluster2_kmeans['Unnamed: 0'].to_list()
cluster3_gene = cluster3_kmeans['Unnamed: 0'].to_list()
cluster4_gene = cluster4_kmeans['Unnamed: 0'].to_list()
cluster5_gene = cluster5_kmeans['Unnamed: 0'].to_list()
cluster6_gene = cluster6_kmeans['Unnamed: 0'].to_list()
cluster7_gene = cluster7_kmeans['Unnamed: 0'].to_list()
cluster8_gene = cluster8_kmeans['Unnamed: 0'].to_list()
cluster9_gene = cluster9_kmeans['Unnamed: 0'].to_list()
cluster10_gene = cluster10_kmeans['Unnamed: 0'].to_list()
cluster11_gene = cluster11_kmeans['Unnamed: 0'].to_list()
cluster12_gene = cluster12_kmeans['Unnamed: 0'].to_list()
cluster13_gene = cluster13_kmeans['Unnamed: 0'].to_list()
cluster14_gene = cluster14_kmeans['Unnamed: 0'].to_list()
In [16]:
clusters_genes = [cluster0_gene, cluster1_gene, cluster2_gene,
            cluster3_gene, cluster4_gene, cluster5_gene, cluster6_gene,
            cluster7_gene, cluster8_gene, cluster9_gene, cluster10_gene,
            cluster11_gene, cluster12_gene, cluster13_gene, cluster14_gene]
In [17]:
kmeans_1_5 = set(cluster1_gene + cluster5_gene)
kmeans_1_5 = list(kmeans_1_5)
len(kmeans_1_5)
Out[17]:
1889
In [18]:
kmeans_2_6 = set(cluster2_gene + cluster6_gene)
kmeans_2_6 = list(kmeans_2_6)
len(kmeans_2_6)
Out[18]:
1871
In [19]:
cluster2_kmeans
Out[19]:
Unnamed: 0 logFC.celltypes_leveledhEGCLC logFC.celltypes_leveledhPGCLC logFC.celltypes_levelediMeLC logCPM LR PValue FDR clusters
0 FST -1.177654 0.328728 5.474972 7.192167 2243.012713 0.000000 0.000000 2
1 VASH1 -1.153742 4.042172 -0.580513 7.137133 1713.609937 0.000000 0.000000 2
2 TFAP2C -1.056494 5.416394 1.400894 8.368850 2019.563791 0.000000 0.000000 2
3 CACNA2D2 -0.692511 4.543920 -1.590436 8.316615 1725.366634 0.000000 0.000000 2
4 SMAGP -0.609917 4.167067 -0.032716 7.181970 1565.755661 0.000000 0.000000 2
... ... ... ... ... ... ... ... ... ...
1494 HIST1H4J 0.364433 0.211762 1.123251 3.989923 15.195154 0.001657 0.002515 2
1495 LYRM2 0.608084 1.075917 0.573123 3.376824 14.442187 0.002361 0.003517 2
1496 PTGES3L 0.759927 1.039564 0.882685 3.358153 14.057661 0.002828 0.004176 2
1497 ZNF780B 0.674144 1.024682 0.727106 3.503380 13.536226 0.003609 0.005251 2
1498 TPRN 0.311433 0.294442 1.156816 3.492064 13.254729 0.004117 0.005950 2

1499 rows × 9 columns

Expected_up = [cluster 2+6 up PGC] & [(low meth PGC vs iPSC) | (low meth PGC vs EG)]¶

In [20]:
LM_PGvsiPSC = DM_hPGCLC_iPSC[DM_hPGCLC_iPSC['diff.MethyhiPSCs-hPGCLCs'] > 0]
LM_PGvsiPSC_list = LM_PGvsiPSC['hgnc_symbol'].unique().tolist()
LM_PGvsiPSC
Out[20]:
seqnames start end width strand length nCG meanMethyhiPSCs meanMethyhPGCLCs diff.MethyhiPSCs-hPGCLCs areaStat ChipSeekerAnn hgnc_symbol gene_biotype
2 chr11 315262 317011 1750 * 1751 93 0.516360 0.177217 0.339143 3136.660612 Promoter (<=1kb) IFITM1 protein_coding
8 chr13 113022412 113023593 1182 * 1183 55 0.767662 0.198155 0.569507 2136.353651 Promoter (1-2kb) MCF2L protein_coding
13 chr17 82875178 82875953 776 * 777 26 0.805230 0.203102 0.602128 908.791067 Intron (ENST00000355528.9/ENSG00000141556.21, ... TBCD protein_coding
14 chr4 6688957 6689447 491 * 492 21 0.746715 0.109023 0.637692 844.883445 Promoter (1-2kb) LINC02481 lncRNA
16 chr21 46140006 46141020 1015 * 1016 30 0.852938 0.382603 0.470335 777.982251 Promoter (1-2kb) FTCD protein_coding
... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
1900 chr6 131061058 131061189 132 * 133 4 0.853553 0.411095 0.442457 40.907887 Promoter (1-2kb) EPB41L2 protein_coding
1901 chrX 106920390 106920475 86 * 87 4 0.637775 0.225380 0.412395 40.618750 Promoter (<=1kb) CLDN2 protein_coding
1902 chr1 41503846 41503976 131 * 132 4 0.734979 0.299798 0.435181 40.211150 Distal Intergenic EDN2 protein_coding
1903 chr1 192983532 192983590 59 * 60 4 0.810811 0.407738 0.403074 40.026621 Distal Intergenic ZNF101P2 processed_pseudogene
1904 chr1 43603067 43603118 52 * 53 4 0.856153 0.410641 0.445513 39.829517 Promoter (2-3kb) PTPRF protein_coding

1834 rows × 14 columns

In [21]:
LM_PGvsEG = DM_hPGCLC_hEGCLC[DM_hPGCLC_hEGCLC['diff.MethyhEGCLCs-hPGCLCs'] > 0]
LM_PGvsEG_list = LM_PGvsEG['hgnc_symbol'].unique().tolist()
LM_PGvsEG
Out[21]:
seqnames start end width strand length nCG meanMethyhEGCLCs meanMethyhPGCLCs diff.MethyhEGCLCs-hPGCLCs areaStat ChipSeekerAnn hgnc_symbol gene_biotype
140 chr13 113022412 113023593 1182 * 1183 55 0.680361 0.198155 0.482206 1853.639233 Promoter (1-2kb) MCF2L protein_coding
296 chr17 82875242 82875953 712 * 713 25 0.824031 0.189483 0.634548 1025.564416 Intron (ENST00000355528.9/ENSG00000141556.21, ... TBCD protein_coding
342 chr4 6688957 6689447 491 * 492 21 0.686484 0.109023 0.577461 801.924394 Promoter (1-2kb) LINC02481 lncRNA
349 chr4 134201320 134201951 632 * 633 50 0.598049 0.221826 0.376223 780.620797 Promoter (<=1kb) PABPC4L protein_coding
363 chr12 121805798 121807062 1265 * 1266 36 0.674091 0.237712 0.436379 724.383849 Promoter (1-2kb) SETD1B protein_coding
... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
2095 chr15 25783669 25783718 50 * 51 4 0.598025 0.121523 0.476503 39.828829 Promoter (<=1kb) ATP10A protein_coding
2096 chr10 86672553 86674114 1562 * 1563 4 0.824302 0.394263 0.430040 37.118959 Intron (ENST00000623056.4/ENSG00000122367.20, ... LDB3 protein_coding
2097 chr8 76810924 76810999 76 * 77 4 0.831456 0.340330 0.491126 36.828297 Intron (ENST00000651372.2/ENSG00000091656.19, ... ZFHX4 protein_coding
2098 chr8 144198100 144198304 205 * 206 4 0.875118 0.439952 0.435166 35.675492 Promoter (<=1kb) MROH1 protein_coding
2099 chr1 62321595 62327534 5940 * 5941 4 0.765698 0.436921 0.328777 20.911213 Promoter (2-3kb) KANK4 protein_coding

1484 rows × 14 columns

In [22]:
LMgenes_PG = set(LM_PGvsiPSC_list + LM_PGvsEG_list)
len(LMgenes_PG)
Out[22]:
1717
In [23]:
expected_up = [gene for gene in kmeans_2_6 if gene in LMgenes_PG]
len(expected_up)
Out[23]:
315

Expected_down = [cluster 1+5 down PGC] & [(high meth PGC vs iPSC) | (high meth PGC vs EG)]¶

In [24]:
HM_PGvsiPSC = DM_hPGCLC_iPSC[DM_hPGCLC_iPSC['diff.MethyhiPSCs-hPGCLCs'] < 0]
HM_PGvsiPSC_list = HM_PGvsiPSC['hgnc_symbol'].unique().tolist()
HM_PGvsiPSC
Out[24]:
seqnames start end width strand length nCG meanMethyhiPSCs meanMethyhPGCLCs diff.MethyhiPSCs-hPGCLCs areaStat ChipSeekerAnn hgnc_symbol gene_biotype
0 chr19 45396154 45397055 902 * 903 99 0.062659 0.510500 -0.447841 -3900.366653 Promoter (2-3kb) PPP1R13L protein_coding
1 chr1 40303423 40304258 836 * 837 61 0.083519 0.746259 -0.662740 -3393.603657 Promoter (<=1kb) COL9A2 protein_coding
3 chr12 6329107 6329910 804 * 805 72 0.239183 0.714567 -0.475384 -2805.981286 Promoter (1-2kb) TNFRSF1A protein_coding
4 chr7 5388141 5388967 827 * 828 82 0.151733 0.539237 -0.387504 -2785.182174 Promoter (<=1kb) TNRC18 protein_coding
5 chr12 114674210 114675406 1197 * 1198 73 0.022812 0.530622 -0.507810 -2689.567789 Exon (ENST00000257566.7/ENSG00000135111.16, ex... TBX3 protein_coding
... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
1724 chr5 72238871 72239091 221 * 222 4 0.076622 0.573147 -0.496525 -50.465300 Promoter (<=1kb) MRPS27 protein_coding
1753 chr13 27963063 27963125 63 * 64 4 0.114860 0.533999 -0.419138 -49.286833 3' UTR CDX2 protein_coding
1800 chr18 686280 686364 85 * 86 4 0.356524 0.759282 -0.402758 -46.848331 Promoter (<=1kb) ENOSF1 protein_coding
1807 chr14 73287178 73287261 84 * 85 4 0.317001 0.794182 -0.477181 -46.566883 Promoter (<=1kb) NUMB protein_coding
1858 chr2 101893983 101894075 93 * 94 4 0.124533 0.567102 -0.442569 -44.279765 3' UTR MAP4K4 protein_coding

71 rows × 14 columns

In [25]:
HM_PGvsEG = DM_hPGCLC_hEGCLC[DM_hPGCLC_hEGCLC['diff.MethyhEGCLCs-hPGCLCs'] < 0]
HM_PGvsEG_list = HM_PGvsEG['hgnc_symbol'].unique().tolist()
HM_PGvsEG
Out[25]:
seqnames start end width strand length nCG meanMethyhEGCLCs meanMethyhPGCLCs diff.MethyhEGCLCs-hPGCLCs areaStat ChipSeekerAnn hgnc_symbol gene_biotype
0 chr1 248847057 248848595 1539 * 1540 152 0.336505 0.938871 -0.602366 -7627.745925 5' UTR ZNF692 protein_coding
1 chr8 38973989 38975246 1258 * 1259 95 0.100581 0.770406 -0.669825 -7260.764787 Promoter (<=1kb) HTRA4 protein_coding
2 chr11 64368176 64370713 2538 * 2539 155 0.283509 0.899267 -0.615758 -7087.566102 Promoter (<=1kb) MIR1237 miRNA
3 chr15 65076644 65077714 1071 * 1072 125 0.431976 0.929163 -0.497187 -6119.045708 Promoter (<=1kb) RASL12 protein_coding
4 chr5 172669729 172670381 653 * 654 89 0.188197 0.736755 -0.548558 -5142.170248 Exon (ENST00000369800.6/ENSG00000214357.9, exo... MIR5003 miRNA
... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
1913 chr7 82508513 82508620 108 * 109 4 0.435300 0.843219 -0.407920 -53.485287 Distal Intergenic CACNA2D1 protein_coding
1992 chrX 51742393 51742737 345 * 346 4 0.104253 0.537215 -0.432961 -49.391265 Promoter (<=1kb) GSPT2 protein_coding
2028 chr6 13422299 13422382 84 * 85 4 0.191870 0.649639 -0.457769 -47.339880 Intron (ENST00000379287.4/ENSG00000145990.11, ... GFOD1 protein_coding
2037 chr4 107906748 107906976 229 * 230 4 0.302545 0.696860 -0.394315 -46.591734 Promoter (<=1kb) SGMS2 protein_coding
2080 chr3 98325132 98325182 51 * 52 4 0.320153 0.736194 -0.416041 -42.934488 Intron (ENST00000508616.1/ENSG00000251088.1, i... OR5H8 polymorphic_pseudogene

616 rows × 14 columns

In [26]:
HMgenes_PG = set(HM_PGvsiPSC_list + HM_PGvsEG_list)
len(HMgenes_PG)
Out[26]:
614
In [27]:
expected_down = [gene for gene in kmeans_1_5 if gene in HMgenes_PG]
len(expected_down)
Out[27]:
36

Expected_all = Expected up | Expected down¶

In [28]:
expected_all = set(expected_up + expected_down)
len(expected_all)
Out[28]:
351

Expected_up_2 = [cluster 2+6 up PGC] | [(low meth PGC vs iPSC) & (low meth PGC vs EG)]¶

In [29]:
LMgenes_PG_union = [gene for gene in LM_PGvsiPSC_list if gene in LM_PGvsEG_list]
len(LMgenes_PG_union)
Out[29]:
1193
In [30]:
expected_up_2 = set(kmeans_2_6 + LMgenes_PG_union)
len(expected_up_2)
Out[30]:
2841

Expected_down_2 = [cluster 1+5 down PGC] | [(high meth PGC vs iPSC) & (high meth PGC vs EG)]¶

In [31]:
HMgenes_PG_union = [gene for gene in HM_PGvsiPSC_list if gene in HM_PGvsEG_list]
len(HMgenes_PG_union)
Out[31]:
42
In [32]:
expected_down_2 = set(kmeans_1_5 + HMgenes_PG_union)
len(expected_down_2)
Out[32]:
1927

Expected_all_2 = Expected up 2 | Expected down 2¶

In [33]:
expected_all_2 = set(list(expected_up_2) + list(expected_down_2))
len(expected_all_2)
Out[33]:
4672

How many genes of the different lists are in the GRNs?¶

In [34]:
target_GRNiPSC = GRN_iPSC['target'].unique()
target_GRNiMeLC = GRN_iMeLC['target'].unique()
target_GRNhPGCLC = GRN_hPGCLC['target'].unique()
target_GRNhEGCLC = GRN_hEGCLC['target'].unique()
In [35]:
print(len(target_GRNiPSC))
print(len(target_GRNiMeLC))
print(len(target_GRNhPGCLC))
print(len(target_GRNhEGCLC))

2895
2895
2895
2895
In [36]:
TF_GRNiPSC = GRN_iPSC['source'].unique()
TF_GRNiMeLC = GRN_iMeLC['source'].unique()
TF_GRNhPGCLC = GRN_hPGCLC['source'].unique()
TF_GRNhEGCLC = GRN_hEGCLC['source'].unique()
In [37]:
print(len(TF_GRNiPSC))
print(len(TF_GRNiMeLC))
print(len(TF_GRNhPGCLC))
print(len(TF_GRNhEGCLC))

255
255
255
255
In [38]:
common_expected_up_GRNiPSC = set(expected_up).intersection(set(target_GRNiPSC))
print('Expected_up targets in GRN {} on {}'.format(len(common_expected_up_GRNiPSC), len(expected_up)))

common_expected_down_GRNiPSC = set(expected_down).intersection(set(target_GRNiPSC))
print('Expected_down targets in GRN {} on {}'.format(len(common_expected_down_GRNiPSC), len(expected_down)))

common_expected_all_GRNiPSC = set(expected_all).intersection(set(target_GRNiPSC))
print('Expected_all targets in GRN {} on {}'.format(len(common_expected_all_GRNiPSC), len(expected_all)))

common_expected_up2_GRNiPSC = set(expected_up_2).intersection(set(target_GRNiPSC))
print('Expected_up2 targets in GRN {} on {}'.format(len(common_expected_up2_GRNiPSC), len(expected_up_2)))

common_expected_down2_GRNiPSC = set(expected_down_2).intersection(set(target_GRNiPSC))
print('Expected_down2 targets in GRN {} on {}'.format(len(common_expected_down2_GRNiPSC), len(expected_down_2)))

common_expected_all2_GRNiPSC = set(expected_all_2).intersection(set(target_GRNiPSC))
print('Expected_all2 targets in GRN {} on {}'.format(len(common_expected_all2_GRNiPSC), len(expected_all_2)))

Expected_up targets in GRN 157 on 315
Expected_down targets in GRN 11 on 36
Expected_all targets in GRN 168 on 351
Expected_up2 targets in GRN 826 on 2841
Expected_down2 targets in GRN 480 on 1927
Expected_all2 targets in GRN 1281 on 4672
In [39]:
commonTF_expected_up_GRNiPSC = set(expected_up).intersection(set(TF_GRNiPSC))
print('Expected_up TFs in GRN {} on {}'.format(len(commonTF_expected_up_GRNiPSC), len(expected_up)))

commonTF_expected_down_GRNiPSC = set(expected_down).intersection(set(TF_GRNiPSC))
print('Expected_down TFs in GRN {} on {}'.format(len(commonTF_expected_down_GRNiPSC), len(expected_down)))

commonTF_expected_all_GRNiPSC = set(expected_all).intersection(set(TF_GRNiPSC))
print('Expected_all TFs in GRN {} on {}'.format(len(commonTF_expected_all_GRNiPSC), len(expected_all)))

commonTF_expected_up2_GRNiPSC = set(expected_up_2).intersection(set(TF_GRNiPSC))
print('Expected_up2 TFs in GRN {} on {}'.format(len(commonTF_expected_up2_GRNiPSC), len(expected_up_2)))

commonTF_expected_down2_GRNiPSC = set(expected_down_2).intersection(set(TF_GRNiPSC))
print('Expected_down2 TFs in GRN {} on {}'.format(len(commonTF_expected_down2_GRNiPSC), len(expected_down_2)))

commonTF_expected_all2_GRNiPSC = set(expected_all_2).intersection(set(TF_GRNiPSC))
print('Expected_all2 TFs in GRN {} on {}'.format(len(commonTF_expected_all2_GRNiPSC), len(expected_all_2)))

Expected_up TFs in GRN 12 on 315
Expected_down TFs in GRN 1 on 36
Expected_all TFs in GRN 13 on 351
Expected_up2 TFs in GRN 69 on 2841
Expected_down2 TFs in GRN 29 on 1927
Expected_all2 TFs in GRN 95 on 4672

Hypergeom and enrichment¶

In [40]:
TF_iPSC = GRN_iPSC_filt['source'].tolist()
TF_iPSC = list(set(TF_iPSC))

TF_iMeLC = GRN_iMeLC_filt['source'].tolist()
TF_iMeLC = list(set(TF_iMeLC))

TF_hPGCLC = GRN_PGCLC_filt['source'].tolist()
TF_hPGCLC = list(set(TF_hPGCLC))

TF_hEGCLC = GRN_EGCLC_filt['source'].tolist()
TF_hEGCLC = list(set(TF_hEGCLC))
In [41]:
def compute_enrichment(TFs, GRN_filt, universo, DMDEs, p_value_threshold):
    
    enrichment_res = []

    for tf in TFs:
        group = GRN_filt[GRN_filt['source'] == tf]
        target_tot = len(group['target'])
        target_DMDE = len(group[group['target'].isin(DMDEs)])
        
        pValueTFx = hypergeom.sf(target_DMDE-1, len(universo), target_tot, len(DMDEs))

        GRN_TF_expressedGenes = group[group['target'].isin(universo)]
        n_targets_expressed = len(GRN_TF_expressedGenes)
        
        GRN_tf_DMDE = group[group['target'].isin(DMDEs)]
        n_targets_DMDE = len(GRN_tf_DMDE)

        if n_targets_expressed == 0:
            ratioTargets = float('nan')
        else:
            ratioTargets = n_targets_DMDE / n_targets_expressed
        
        ratioDMDE = len(DMDEs) / len(universo)
        
        enrichment = ratioTargets / ratioDMDE
        
        enrichment_res.append({
            'TF': tf,
            'targetTF': target_tot,
            'targetDMDE': target_DMDE,
            'pValue': pValueTFx,
            'enrichment': enrichment
        })

    enrichment_df = pd.DataFrame(enrichment_res)
    significant_enrichment_df = enrichment_df[enrichment_df['pValue'] < p_value_threshold]
    
    return significant_enrichment_df
In [42]:
universo = set()

for cluster in clusters_genes:
    universo.update(cluster)

universo.update(tot_geni_DM_PGvsiPSC)
universo.update(tot_geni_DM_PGvsEG)
universo.update(tot_geni_DM_iPSCvsEG)
In [43]:
len(universo)
Out[43]:
6659
In [44]:
expected_up_enrichment_iPSC = compute_enrichment(TF_iPSC, GRN_iPSC_filt, universo, expected_up, p_value_threshold=0.05)
print(expected_up_enrichment_iPSC.sort_values(by='enrichment', ascending=False))
expected_up_enrichment_iPSC.to_csv('expected_up_enrichment_iPSC.csv', index=False)

expected_up_enrichment_iMeLC = compute_enrichment(TF_iMeLC, GRN_iMeLC_filt, universo, expected_up, p_value_threshold=0.05)
print(expected_up_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))
expected_up_enrichment_iMeLC.to_csv('expected_up_enrichment_iMeLC.csv', index=False)

expected_up_enrichment_PG = compute_enrichment(TF_hPGCLC, GRN_PGCLC_filt, universo, expected_up, p_value_threshold=0.05)
print(expected_up_enrichment_PG.sort_values(by='enrichment', ascending=False))
expected_up_enrichment_PG.to_csv('expected_up_enrichment_PG.csv', index=False)

expected_up_enrichment_EG = compute_enrichment(TF_hEGCLC, GRN_EGCLC_filt, universo, expected_up, p_value_threshold=0.05)
print(expected_up_enrichment_EG.sort_values(by='enrichment', ascending=False))
expected_up_enrichment_EG.to_csv('expected_up_enrichment_EG.csv', index=False)

          TF  targetTF  targetDMDE    pValue  enrichment
80    PKNOX2        49           6  0.026928    5.514700
1       TBX3        25           4  0.028308    4.697707
77      GBX2       149          16  0.001716    3.979234
203     LMO2       243          26  0.000072    3.639945
48     FOXP1       457          38  0.000403    2.931781
113    ATOH8       109          10  0.032884    2.895847
129   TCF7L1       301          22  0.027174    2.784868
164    MEF2B       235          18  0.029051    2.737513
83      BCL6       211          16  0.041336    2.642460
165   PRDM11       215          16  0.047692    2.562386
91     NR2F2       287          21  0.030173    2.522348
142    NR5A2       393          28  0.018647    2.508098
15     FOXB1       299          21  0.044033    2.466296
157    GLIS3       646          46  0.002817    2.377568
152     E2F7       730          47  0.016227    2.360012
90     FOSL2       328          24  0.021337    2.359779
86      E2F2       555          38  0.012227    2.355742
53    NFATC4       428          29  0.030634    2.348854
186    NFKB2       722          49  0.005362    2.296773
40     FOXO1       500          32  0.047208    2.254899
78      SOX2       588          39  0.018110    2.252589
197     ZIC5       874          57  0.006240    2.252265
210     ZIC2       852          55  0.008894    2.231636
215     IRX2       355          24  0.047723    2.225230
220      MAF       425          28  0.045195    2.208624
28     SOX21       564          38  0.015561    2.176986
162  BHLHE40       733          47  0.017377    2.159924
4     ZNF165       670          42  0.033544    2.149798
154     REST      1082          64  0.029449    2.137345
182    FOXP4       766          49  0.015850    2.122632
134      ID2       802          48  0.048211    2.113968
175    HDAC2      1306          75  0.034352    2.072518
145    CTCFL       851          51  0.041395    2.022746
59      EGR1      1075          62  0.049994    1.950387
          TF  targetTF  targetDMDE    pValue  enrichment
205    PITX2        44           6  0.016533    4.878388
100    PPARA       178          21  0.000090    3.730532
117     EBF1        87           9  0.021169    3.397449
47     ZFHX3       150          15  0.004702    3.269023
140     EGR2       163          17  0.001713    3.098057
178    FOXO4       380          31  0.001904    2.861704
192    SNAI1       137          12  0.028340    2.850294
195     LMO2       237          19  0.016520    2.751055
45     FOXP1       394          30  0.006192    2.710216
87     FOSL2       217          18  0.014082    2.698683
179    NFKB2       385          30  0.004452    2.687248
81      ZIC3       426          33  0.003169    2.612770
37     FOXO1       547          41  0.001896    2.602784
26     TFCP2       830          56  0.003260    2.584765
93     KLF15       532          41  0.001112    2.579545
153  BHLHE40       561          47  0.000065    2.574003
144     E2F7       672          49  0.001226    2.557641
190    NPAS1       329          25  0.012329    2.553102
19     SOX18       159          13  0.037600    2.544591
82      E2F2       352          25  0.026355    2.528670
122   TCF7L1       265          19  0.045265    2.526126
24      PBX3       929          62  0.002483    2.515663
198    STAT2       742          51  0.003466    2.507265
145     HES1       644          46  0.002650    2.506251
155     UBP1       441          30  0.026855    2.477307
186     IRX4       175          15  0.018280    2.477307
108    FOXH1       284          20  0.047519    2.472477
141     HEY2       553          35  0.044672    2.466296
3     ZNF165       611          42  0.007991    2.439194
68     PRDM1       408          32  0.003015    2.390353
119    ESRRB       318          22  0.045869    2.302342
142     KLF5       785          56  0.000899    2.298684
44      MAFB       477          31  0.042364    2.291364
210    FOXI3       702          51  0.001047    2.264966
61    TFAP2C       627          44  0.004579    2.263129
112    OVOL2       672          43  0.023339    2.250016
25     SOX21       261          19  0.039776    2.243877
39       SP3      1068          64  0.022936    2.243681
16      ETS2       587          37  0.041714    2.241170
196    FOXA2       283          22  0.014405    2.235928
146     REST       929          61  0.003950    2.200547
22      ATF3       617          40  0.023503    2.184980
202     ZIC2       670          43  0.022305    2.169466
33     FOSL1       555          36  0.030727    2.149798
7       ETV5       557          35  0.048725    2.113968
168    HDAC2      1266          72  0.046032    2.085010
171    FOXD3       584          39  0.016366    2.055979
56      EGR1       942          56  0.037879    2.051685
189     ZIC5       851          51  0.041395    1.992835
         TF  targetTF  targetDMDE        pValue  enrichment
29   ZNF146        23           4  2.133171e-02    7.046561
13   ZNF232        12           3  1.677383e-02    7.046561
130   ASCL1        94          21  1.572785e-09    6.081279
120   PRRX2        61           7  2.416122e-02    5.480658
90      ZFY        87          10  7.644030e-03    5.420431
..      ...       ...         ...           ...         ...
113   NR5A1        86           8  4.972504e-02    2.818624
84    MEF2A       578          42  2.992472e-03    2.809705
157   FOXB2       309          25  5.733376e-03    2.724186
158  TCF7L2       324          30  2.838842e-04    2.687248
75     BCL6        90           9  2.579248e-02    2.679678

[101 rows x 5 columns]
         TF  targetTF  targetDMDE    pValue  enrichment
174    DLX5        66           7  0.035367    3.999399
31    ARNT2       164          13  0.046388    2.862665
202    LMO2       336          27  0.004552    2.744093
93    NR2F2       385          28  0.014512    2.584765
185   FOXO4       564          39  0.009571    2.439194
64   ARID5B       326          24  0.019973    2.415964
131   NR5A1       399          29  0.013070    2.394730
168    FOSB       502          34  0.020202    2.387871
157   GLIS3       663          47  0.002765    2.376950
177    KLF6      1040          68  0.002511    2.360424
101    EBF3       481          31  0.046555    2.357303
72    PRDM1       580          41  0.005460    2.342505
18     ETS2       933          59  0.010291    2.335658
48    FOXP1       495          33  0.026920    2.309965
143   NR5A2       389          26  0.045374    2.309377
25     ATF3       849          58  0.002047    2.304702
210     MYB       462          30  0.045820    2.273084
29    TFCP2       851          52  0.028919    2.247983
52   NFATC4       595          37  0.049417    2.178742
3     GTF2I       811          49  0.040035    2.171582
110    JUNB       706          43  0.047482    2.169466
122   OVOL2       719          44  0.042320    2.163130
85     ZIC3       853          54  0.013844    2.162013
204   ZFP57       800          49  0.032427    2.122632
205   STAT2       859          53  0.023293    2.121976
155     MYC      1107          66  0.023131    2.107582
80      FOS      1106          66  0.022718    2.088651
67      ID1       869          52  0.040543    2.081938
27     PBX3      1062          64  0.020529    2.062408
195    ZIC5       926          55  0.040110    2.050586
145   CTCFL       702          44  0.029940    2.039794
117     JUN       812          49  0.040790    2.027093
60     EGR1      1107          65  0.032436    1.991419
In [45]:
expected_up2_enrichment_iPSC = compute_enrichment(TF_iPSC, GRN_iPSC_filt, universo, expected_up_2, p_value_threshold=0.05)
print(expected_up2_enrichment_iPSC.sort_values(by='enrichment', ascending=False))

expected_up2_enrichment_iMeLC = compute_enrichment(TF_iMeLC, GRN_iMeLC_filt, universo, expected_up_2, p_value_threshold=0.05)
print(expected_up2_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))

expected_up2_enrichment_PG = compute_enrichment(TF_hPGCLC, GRN_PGCLC_filt, universo, expected_up_2, p_value_threshold=0.05)
print(expected_up2_enrichment_PG.sort_values(by='enrichment', ascending=False))

expected_up2_enrichment_EG = compute_enrichment(TF_hEGCLC, GRN_EGCLC_filt, universo, expected_up_2, p_value_threshold=0.05)
print(expected_up2_enrichment_EG.sort_values(by='enrichment', ascending=False))

Empty DataFrame
Columns: [TF, targetTF, targetDMDE, pValue, enrichment]
Index: []
Empty DataFrame
Columns: [TF, targetTF, targetDMDE, pValue, enrichment]
Index: []
         TF  targetTF  targetDMDE        pValue  enrichment
117   HOXC5        14          10  2.855208e-02    2.343893
13   ZNF232        12           9  2.427715e-02    2.343893
95     JUNB        91          64  7.300515e-08    2.054920
165    SOX7        18          13  1.083850e-02    2.031374
118   DDIT3        79          45  7.029497e-03    2.028369
..      ...       ...         ...           ...         ...
42   TFAP2A       405         206  3.698634e-04    1.631223
22     ATF3       236         117  1.738710e-02    1.576066
62      ID1       295         165  1.892897e-06    1.565758
155   FOXD3        61          35  1.420525e-02    1.491568
184    ZIC2       124          63  3.985584e-02    1.476653

[68 rows x 5 columns]
Empty DataFrame
Columns: [TF, targetTF, targetDMDE, pValue, enrichment]
Index: []
In [46]:
expected_down_enrichment_iPSC = compute_enrichment(TF_iPSC, GRN_iPSC_filt, universo, expected_down, p_value_threshold=0.05)
print(expected_down_enrichment_iPSC.sort_values(by='enrichment', ascending=False))
expected_down_enrichment_iPSC.to_csv('expected_down_enrichment_iPSC.csv', index=False)

expected_down_enrichment_iMeLC = compute_enrichment(TF_iMeLC, GRN_iMeLC_filt, universo, expected_down, p_value_threshold=0.05)
print(expected_down_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))
expected_down_enrichment_iMeLC.to_csv('expected_down_enrichment_iMeLC.csv', index=False)

expected_down_enrichment_PG = compute_enrichment(TF_hPGCLC, GRN_PGCLC_filt, universo, expected_down, p_value_threshold=0.05)
print(expected_down_enrichment_PG.sort_values(by='enrichment', ascending=False))
expected_down_enrichment_PG.to_csv('expected_down_enrichment_PG.csv', index=False)

expected_down_enrichment_EG = compute_enrichment(TF_hEGCLC, GRN_EGCLC_filt, universo, expected_down, p_value_threshold=0.05)
print(expected_down_enrichment_EG.sort_values(by='enrichment', ascending=False))
expected_down_enrichment_EG.to_csv('expected_down_enrichment_EG.csv', index=False)

          TF  targetTF  targetDMDE    pValue  enrichment
208   POU3F4        27           2  0.009163   23.121528
147     IRX3        45           2  0.024321   15.414352
73      IRX5       170           4  0.012779    8.407828
189   ZNF816       151           4  0.008515    7.871158
165   PRDM11       215           4  0.027745    5.605219
7      ZBTB6       231           4  0.034858    5.563074
201  ONECUT1       250           4  0.044581    4.773477
218  CREB3L1       646           8  0.019635    3.699444
40     FOXO1       500           6  0.049310    3.699444
4     ZNF165       670           8  0.023960    3.582997
118     E2F1       900           9  0.045840    3.251465
          TF  targetTF  targetDMDE    pValue  enrichment
211  CREB3L1       580           8  0.010686    4.043109
      TF  targetTF  targetDMDE    pValue  enrichment
56  CDX2        61           2  0.042609    7.707176
          TF  targetTF  targetDMDE    pValue  enrichment
169    SNAI2        62           3  0.004410   15.854762
208   ZNF136        41           2  0.020418   15.414352
7      ZBTB6       238           5  0.008522    7.282371
199  ONECUT1       309           5  0.024089    4.718679
187     ATF5       328           5  0.030212    4.533633
84      BCL6       335           5  0.032697    3.986470
184    PPARG       495           6  0.047328    3.774943
In [47]:
expected_down2_enrichment_iPSC = compute_enrichment(TF_iPSC, GRN_iPSC_filt, universo, expected_down_2, p_value_threshold=0.05)
print(expected_down2_enrichment_iPSC.sort_values(by='enrichment', ascending=False))

expected_down2_enrichment_iMeLC = compute_enrichment(TF_iMeLC, GRN_iMeLC_filt, universo, expected_down_2, p_value_threshold=0.05)
print(expected_down2_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))

expected_down2_enrichment_PG = compute_enrichment(TF_hPGCLC, GRN_PGCLC_filt, universo, expected_down_2, p_value_threshold=0.05)
print(expected_down2_enrichment_PG.sort_values(by='enrichment', ascending=False))

expected_down2_enrichment_EG = compute_enrichment(TF_hEGCLC, GRN_EGCLC_filt, universo, expected_down_2, p_value_threshold=0.05)
print(expected_down2_enrichment_EG.sort_values(by='enrichment', ascending=False))

          TF  targetTF  targetDMDE    pValue  enrichment
8     ZNF467        33          17  0.005154    2.670260
143     TP63        45          22  0.003590    2.452383
125     EBF1        47          21  0.015340    2.267758
85     HOXB6        68          29  0.010570    2.178550
106    PPARA       126          51  0.003346    2.098061
137    PRRX2       124          46  0.029279    2.091566
189   ZNF816       151          56  0.017625    2.058673
33    POU4F1       174          61  0.044598    2.026860
72     INSM1       141          54  0.009792    2.006495
122    BARX1       147          58  0.003621    2.004266
17   BHLHE41       196          76  0.001679    1.989605
113    ATOH8       109          41  0.030355    1.940834
67     HOXB2       248          85  0.036081    1.932425
30       VDR       272          97  0.008404    1.841737
90     FOSL2       328         112  0.020309    1.800142
102      SP6       436         143  0.038320    1.727815
        TF  targetTF  targetDMDE    pValue  enrichment
106  ATOH8        20          10  0.037712    2.468308
103  GATA5        18           9  0.047937    2.221477
12   FOXB1       223          79  0.019286    1.727815
      TF  targetTF  targetDMDE    pValue  enrichment
70  SOX2       109          41  0.030355    1.460627
        TF  targetTF  targetDMDE    pValue  enrichment
51   FOXD1        11           7  0.017561    2.687713
147   IRX3        37          18  0.008661    2.488054
89    DLX2        43          18  0.047321    2.303754
137  PRRX2       124          45  0.044595    2.221477
212  PITX2        36          16  0.033987    2.126542
129  ESRRB        68          30  0.005277    2.115692
11    MITF        62          25  0.035295    2.107092
206  FOXA2       121          54  0.000155    1.964253
30     VDR       276          95  0.024959    1.897601
84    BCL6       335         111  0.048139    1.653340
In [48]:
expected_all_enrichment_iPSC = compute_enrichment(TF_iPSC, GRN_iPSC_filt, universo, expected_all, p_value_threshold=0.05)
print(expected_all_enrichment_iPSC.sort_values(by='enrichment', ascending=False))
expected_all_enrichment_iPSC.to_csv('expected_all_enrichment_iPSC.csv', index=False)

expected_all_enrichment_iMeLC = compute_enrichment(TF_iMeLC, GRN_iMeLC_filt, universo, expected_all, p_value_threshold=0.05)
print(expected_all_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))
expected_all_enrichment_iMeLC.to_csv('expected_all_enrichment_iMeLC.csv', index=False)

expected_all_enrichment_PG = compute_enrichment(TF_hPGCLC, GRN_PGCLC_filt, universo, expected_all, p_value_threshold=0.05)
print(expected_all_enrichment_PG.sort_values(by='enrichment', ascending=False))
expected_all_enrichment_PG.to_csv('expected_all_enrichment_PG.csv', index=False)

expected_all_enrichment_EG = compute_enrichment(TF_hEGCLC, GRN_EGCLC_filt, universo, expected_all, p_value_threshold=0.05)
print(expected_all_enrichment_EG.sort_values(by='enrichment', ascending=False))
expected_all_enrichment_EG.to_csv('expected_all_enrichment_EG.csv', index=False)

          TF  targetTF  targetDMDE    pValue  enrichment
51     FOXD1        14           3  0.034260    5.691453
80    PKNOX2        49           6  0.042459    4.949090
147     IRX3        45           6  0.029433    4.742877
1       TBX3        25           4  0.039921    4.215891
213    PITX2        45           6  0.029433    3.794302
77      GBX2       149          16  0.005044    3.571108
203     LMO2       243          27  0.000172    3.392257
48     FOXP1       457          42  0.000239    2.908042
165   PRDM11       215          20  0.009224    2.874471
164    MEF2B       235          21  0.011954    2.866199
113    ATOH8       109          11  0.027971    2.858721
129   TCF7L1       301          24  0.027010    2.726445
83      BCL6       211          18  0.029040    2.667869
91     NR2F2       287          23  0.028521    2.479231
157    GLIS3       646          53  0.000627    2.458411
40     FOXO1       500          38  0.013110    2.403058
86      E2F2       555          43  0.006082    2.392302
90     FOSL2       328          27  0.013458    2.382469
99      EBF3       446          32  0.044243    2.343970
142    NR5A2       393          29  0.039774    2.331245
185    FOXO4       562          41  0.019180    2.328838
120    OVOL2       766          52  0.031013    2.321220
152     E2F7       730          51  0.020229    2.298211
4     ZNF165       670          50  0.006538    2.296793
53    NFATC4       428          31  0.042456    2.253321
215     IRX2       355          27  0.033552    2.246626
186    NFKB2       722          53  0.007114    2.229468
78      SOX2       588          43  0.015997    2.228893
168     FOSB       520          37  0.035598    2.200457
220      MAF       425          31  0.039170    2.194466
210     ZIC2       852          60  0.010148    2.184819
197     ZIC5       874          61  0.011482    2.163107
153     HES1       584          40  0.049286    2.161995
28     SOX21       564          42  0.012920    2.159359
154     REST      1082          72  0.017747    2.157897
218  CREB3L1       646          45  0.029838    2.134295
182    FOXP4       766          54  0.014406    2.099306
134      ID2       802          53  0.045522    2.094771
42       SP3      1255          79  0.043815    2.081596
162  BHLHE40       733          50  0.031672    2.062121
175    HDAC2      1306          82  0.042242    2.033547
145    CTCFL       851          57  0.030752    2.028848
181   TCF7L2      1206          77  0.034969    1.971398
180    FOXB2       716          48  0.045354    1.966809
59      EGR1      1075          69  0.041309    1.947968
          TF  targetTF  targetDMDE    pValue  enrichment
205    PITX2        44           6  0.026647    4.378041
100    PPARA       178          22  0.000151    3.507338
47     ZFHX3       150          16  0.005382    3.129321
117     EBF1        87           9  0.038807    3.048993
45     FOXP1       394          35  0.001473    2.837619
140     EGR2       163          17  0.005253    2.780307
98      LHX1       131          12  0.042853    2.776319
178    FOXO4       380          33  0.002896    2.733886
123     TCF7       222          18  0.044615    2.667869
122   TCF7L1       265          22  0.022370    2.624989
121    NR5A1       185          16  0.034286    2.616760
48      RARB       163          15  0.024517    2.610758
195     LMO2       237          20  0.024541    2.598837
179    NFKB2       385          32  0.006462    2.572408
26     TFCP2       830          62  0.002352    2.568196
87     FOSL2       217          19  0.019892    2.556445
37     FOXO1       547          44  0.002755    2.506746
145     HES1       644          51  0.001767    2.493678
198    STAT2       742          56  0.003119    2.470708
3     ZNF165       611          47  0.004760    2.449618
81      ZIC3       426          34  0.009300    2.415848
24      PBX3       929          66  0.005688    2.403301
144     E2F7       672          51  0.004311    2.389005
190    NPAS1       329          26  0.024176    2.382895
155     UBP1       441          32  0.038802    2.371439
93     KLF15       532          42  0.004818    2.371439
82      E2F2       352          26  0.049373    2.360092
153  BHLHE40       561          48  0.000456    2.359151
119    ESRRB       318          25  0.028293    2.347959
68     PRDM1       408          34  0.004849    2.279263
210    FOXI3       702          56  0.000866    2.231942
25     SOX21       261          21  0.034102    2.225708
186     IRX4       175          15  0.042344    2.223224
142     KLF5       785          60  0.001617    2.210273
112    OVOL2       672          47  0.025025    2.207082
61    TFAP2C       627          47  0.007699    2.169491
33     FOSL1       555          40  0.024589    2.143673
39       SP3      1068          68  0.049468    2.139407
146     REST       929          66  0.005688    2.136723
202     ZIC2       670          47  0.023869    2.128069
7       ETV5       557          39  0.038999    2.113968
22      ATF3       617          43  0.033153    2.107946
56      EGR1       942          62  0.033830    2.038533
196    FOXA2       283          22  0.042406    2.006602
171    FOXD3       584          42  0.022198    1.987041
189     ZIC5       851          56  0.043164    1.963779
         TF  targetTF  targetDMDE        pValue  enrichment
29   ZNF146        23           4  3.033423e-02    6.323837
13   ZNF232        12           3  2.238388e-02    6.323837
130   ASCL1        94          21  1.104807e-08    5.457558
41    FOXP1       158          31  1.078574e-10    5.026639
8     EPAS1        43           7  6.634368e-03    4.918540
..      ...       ...         ...           ...         ...
158  TCF7L2       324          33  1.787475e-04    2.652796
157   FOXB2       309          27  6.305937e-03    2.640365
94    PPARA       208          17  4.701346e-02    2.600933
84    MEF2A       578          42  1.897949e-02    2.521530
75     BCL6        90           9  4.664299e-02    2.404839

[101 rows x 5 columns]
          TF  targetTF  targetDMDE    pValue  enrichment
212    PITX2        36           5  0.038857    3.648367
91      RXRG       225          19  0.027804    2.838257
202     LMO2       336          30  0.002957    2.736275
132   TCF7L1       322          25  0.032278    2.649652
93     NR2F2       385          31  0.011342    2.568196
185    FOXO4       564          45  0.002887    2.525793
26    ZNF394       354          27  0.032537    2.523304
131    NR5A1       399          33  0.006152    2.445546
49     ZFHX3       255          20  0.047627    2.401457
168     FOSB       502          38  0.013920    2.395074
48     FOXP1       495          38  0.011253    2.387144
72     PRDM1       580          46  0.002915    2.358620
64    ARID5B       326          26  0.021818    2.348854
177     KLF6      1040          75  0.002057    2.336393
22     SOX18       244          20  0.032189    2.327793
199  ONECUT1       309          24  0.035346    2.323042
101     EBF3       481          34  0.046455    2.320257
18      ETS2       933          65  0.009410    2.309266
182    FOXP4       720          50  0.023691    2.285724
157    GLIS3       663          50  0.005334    2.269319
29     TFCP2       851          58  0.021490    2.250200
204    ZFP57       800          57  0.009700    2.215935
25      ATF3       849          62  0.004006    2.210965
84      BCL6       335          27  0.017354    2.207891
122    OVOL2       719          50  0.023151    2.205990
215     IRX2       331          25  0.042789    2.195777
85      ZIC3       853          61  0.006876    2.191784
110     JUNB       706          48  0.036786    2.173347
205    STAT2       859          60  0.011990    2.155853
67       ID1       869          60  0.015101    2.155853
79      SOX2       603          42  0.035351    2.130490
145    CTCFL       702          51  0.010059    2.121814
155      MYC      1107          74  0.014629    2.120682
80       FOS      1106          74  0.014333    2.101634
218  CREB3L1       646          45  0.029838    2.072131
161  BHLHE40       949          62  0.038594    2.045624
217    FOXI3       872          57  0.046269    2.040332
60      EGR1      1107          73  0.020640    2.007131
117      JUN       812          54  0.039525    2.004817
27      PBX3      1062          69  0.032786    1.995479
In [49]:
expected_all2_enrichment_iPSC = compute_enrichment(TF_iPSC, GRN_iPSC_filt, universo, expected_all_2, p_value_threshold=0.05)
print(expected_all2_enrichment_iPSC.sort_values(by='enrichment', ascending=False))

expected_all2_enrichment_iMeLC = compute_enrichment(TF_iMeLC, GRN_iMeLC_filt, universo, expected_all_2, p_value_threshold=0.05)
print(expected_all2_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))

expected_all2_enrichment_PG = compute_enrichment(TF_hPGCLC, GRN_PGCLC_filt, universo, expected_all_2, p_value_threshold=0.05)
print(expected_all2_enrichment_PG.sort_values(by='enrichment', ascending=False))

expected_all2_enrichment_EG = compute_enrichment(TF_hEGCLC, GRN_EGCLC_filt, universo, expected_all_2, p_value_threshold=0.05)
print(expected_all2_enrichment_EG.sort_values(by='enrichment', ascending=False))

Empty DataFrame
Columns: [TF, targetTF, targetDMDE, pValue, enrichment]
Index: []
Empty DataFrame
Columns: [TF, targetTF, targetDMDE, pValue, enrichment]
Index: []
          TF  targetTF  targetDMDE        pValue  enrichment
141    MEF2B        20          18  3.644723e-02    1.425300
70      SOX2       109          94  6.362938e-05    1.381218
171     ZIC5       256         192  4.717329e-02    1.375163
155    FOXD3        61          53  1.843259e-03    1.373471
102  TFCP2L1       153         127  1.663723e-04    1.371311
76      ZIC3        43          36  3.230007e-02    1.350284
189    FOXI3       300         253  4.806325e-09    1.340523
20     OVOL1        94          76  1.262683e-02    1.337318
62       ID1       295         231  8.388300e-04    1.332973
60    TFAP2C       539         436  2.514092e-09    1.322193
143    SOX17       147         122  2.263843e-04    1.317322
2       KLF4       492         399  6.964396e-09    1.316423
50      MSX2       106          95  1.286136e-06    1.314597
133     REST       450         337  1.242489e-02    1.312366
119      ID2       182         143  6.314212e-03    1.306525
86     KLF15       103          84  5.729035e-03    1.287367
Empty DataFrame
Columns: [TF, targetTF, targetDMDE, pValue, enrichment]
Index: []
In [50]:
def plot_enrichment(df, enrichment_threshold, size_col='targetDMDE', figsize=(8, 6)):

    enrichment_filt = df[df['enrichment'] > enrichment_threshold]

    enrichment_filt = enrichment_filt.sort_values(by='enrichment', ascending=False)

    enrichment_filt['TF'] = pd.Categorical(enrichment_filt['TF'], categories=enrichment_filt['TF'].unique(), ordered=True)

    plt.figure(figsize=figsize)

    dotplot = sns.scatterplot(
        x='enrichment', 
        y='TF',
        hue='pValue', 
        size=size_col,
        data=enrichment_filt,
        palette='viridis',
        sizes=(50, 200),
        legend=True
    )

    plt.xlabel('Enrichment')
    plt.ylabel('Transcription Factors (TF)')
    plt.legend(loc='lower right', prop={'size': 10})
    plt.show()
In [51]:
plot_enrichment(expected_up_enrichment_iPSC, enrichment_threshold=1)
plot_enrichment(expected_up_enrichment_iMeLC, enrichment_threshold=2.3)
plot_enrichment(expected_up_enrichment_PG, enrichment_threshold=4.2)
plot_enrichment(expected_up_enrichment_EG, enrichment_threshold=1)
In [52]:
plot_enrichment(expected_down_enrichment_iPSC, enrichment_threshold=1)
plot_enrichment(expected_down_enrichment_iMeLC, enrichment_threshold=1)
plot_enrichment(expected_down_enrichment_PG, enrichment_threshold=1)
plot_enrichment(expected_down_enrichment_EG, enrichment_threshold=1)
In [53]:
plot_enrichment(expected_all_enrichment_iPSC, enrichment_threshold=2.2)
plot_enrichment(expected_all_enrichment_iMeLC, enrichment_threshold=2.3)
plot_enrichment(expected_all_enrichment_PG, enrichment_threshold=4)
plot_enrichment(expected_all_enrichment_EG, enrichment_threshold=2.2)
In [54]:
plot_enrichment(expected_up2_enrichment_iPSC, enrichment_threshold=1)
plot_enrichment(expected_up2_enrichment_iMeLC, enrichment_threshold=1)
plot_enrichment(expected_up2_enrichment_PG, enrichment_threshold=1.9)
plot_enrichment(expected_up2_enrichment_EG, enrichment_threshold=1)

/tmp/ipykernel_3576080/4072407312.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/4072407312.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/4072407312.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.
In [55]:
plot_enrichment(expected_down2_enrichment_iPSC, enrichment_threshold=1)
plot_enrichment(expected_down2_enrichment_iMeLC, enrichment_threshold=1)
plot_enrichment(expected_down2_enrichment_PG, enrichment_threshold=1)
plot_enrichment(expected_down2_enrichment_EG, enrichment_threshold=1)
In [56]:
plot_enrichment(expected_all2_enrichment_iPSC, enrichment_threshold=1)
plot_enrichment(expected_all2_enrichment_iMeLC, enrichment_threshold=1)
plot_enrichment(expected_all2_enrichment_PG, enrichment_threshold=1)
plot_enrichment(expected_all2_enrichment_EG, enrichment_threshold=1)

/tmp/ipykernel_3576080/4072407312.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/4072407312.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/4072407312.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

Sorted by pvlaue¶

In [57]:
def plot_pvalue(df, pvalue_threshold, size_col='targetDMDE', figsize=(8, 6)):

    enrichment_filt = df[df['pValue'] < pvalue_threshold]

    enrichment_filt = enrichment_filt.sort_values(by='pValue', ascending=True)

    enrichment_filt['TF'] = pd.Categorical(enrichment_filt['TF'], categories=enrichment_filt['TF'].unique(), ordered=True)

    plt.figure(figsize=figsize)

    dotplot = sns.scatterplot(
        x='pValue', 
        y='TF',
        hue='enrichment', 
        size=size_col,
        data=enrichment_filt,
        palette='viridis',
        sizes=(50, 200),
        legend=True
    )

    plt.xlabel('pValue')
    plt.ylabel('Transcription Factors (TF)')
    plt.legend(loc='upper right', prop={'size': 10})
    plt.show()
In [58]:
plot_pvalue(expected_up_enrichment_iPSC, pvalue_threshold=0.04)
plot_pvalue(expected_up_enrichment_iMeLC, pvalue_threshold=0.025)
plot_pvalue(expected_up_enrichment_PG, pvalue_threshold=0.0000015)
plot_pvalue(expected_up_enrichment_EG, pvalue_threshold=0.045)
In [59]:
plot_pvalue(expected_down_enrichment_iPSC, pvalue_threshold=1)
plot_pvalue(expected_down_enrichment_iMeLC, pvalue_threshold=1)
plot_pvalue(expected_down_enrichment_PG, pvalue_threshold=1)
plot_pvalue(expected_down_enrichment_EG, pvalue_threshold=1)
In [60]:
plot_pvalue(expected_all_enrichment_iPSC, pvalue_threshold=0.035)
plot_pvalue(expected_all_enrichment_iMeLC, pvalue_threshold=0.025)
plot_pvalue(expected_all_enrichment_PG, pvalue_threshold=0.0000015)
plot_pvalue(expected_all_enrichment_EG, pvalue_threshold=0.035)
In [61]:
plot_pvalue(expected_up2_enrichment_iPSC, pvalue_threshold=0.05)
plot_pvalue(expected_up2_enrichment_iMeLC, pvalue_threshold=0.05)
plot_pvalue(expected_up2_enrichment_PG, pvalue_threshold=0.00001)
plot_pvalue(expected_up2_enrichment_EG, pvalue_threshold=0.05)

/tmp/ipykernel_3576080/3428210524.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/3428210524.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/3428210524.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.
In [62]:
plot_pvalue(expected_down2_enrichment_iPSC, pvalue_threshold=1)
plot_pvalue(expected_down2_enrichment_iMeLC, pvalue_threshold=1)
plot_pvalue(expected_down2_enrichment_PG, pvalue_threshold=1)
plot_pvalue(expected_down2_enrichment_EG, pvalue_threshold=1)
In [63]:
plot_pvalue(expected_all2_enrichment_iPSC, pvalue_threshold=0.05)
plot_pvalue(expected_all2_enrichment_iMeLC, pvalue_threshold=0.05)
plot_pvalue(expected_all2_enrichment_PG, pvalue_threshold=0.05)
plot_pvalue(expected_all2_enrichment_EG, pvalue_threshold=0.05)

/tmp/ipykernel_3576080/3428210524.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/3428210524.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

/tmp/ipykernel_3576080/3428210524.py:11: UserWarning: Ignoring `palette` because no `hue` variable has been assigned.
  dotplot = sns.scatterplot(
No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

TFs DMDE analysis¶

In [68]:
commonTF_expected_down_GRNiPSC
Out[68]:
{'ZIC5'}
In [69]:
commonTF_expected_all_GRNiPSC
Out[69]:
{'ELF3',
 'ELK3',
 'EZH2',
 'FOXP1',
 'GATA3',
 'LMO2',
 'MSX2',
 'OVOL2',
 'PRDM14',
 'PRRX2',
 'RARB',
 'TCF7',
 'ZIC5'}
In [70]:
TF_dmde_grn = GRN_iPSC[GRN_iPSC['source'].isin(commonTF_expected_all_GRNiPSC)]
TF_dmde_grn
Out[70]:
Unnamed: 0 source target coef_mean coef_abs p -logp
29 29 ELF3 A4GALT 0.013303 0.013303 5.879207e-13 12.230681
53 53 LMO2 A4GALT 0.017735 0.017735 1.102182e-13 12.957747
56 56 OVOL2 A4GALT 0.001903 0.001903 1.500717e-03 2.823701
90 90 ZIC5 A4GALT 0.013395 0.013395 1.373494e-07 6.862173
105 105 GATA3 A4GALT 0.000290 0.000290 4.910657e-01 0.308860
... ... ... ... ... ... ... ...
350371 350371 GATA3 ZSWIM7 -0.006067 0.006067 8.498659e-06 5.070650
350404 350404 EZH2 ZYG11A -0.007923 0.007923 5.477281e-05 4.261435
350427 350427 OVOL2 ZYG11A 0.007120 0.007120 3.897634e-09 8.409199
350456 350456 ZIC5 ZYG11A -0.005095 0.005095 6.957498e-05 4.157547
350468 350468 GATA3 ZYG11A 0.004826 0.004826 1.184023e-07 6.926640

16658 rows × 7 columns

In [71]:
TFdmde_target = TF_dmde_grn.groupby('source')['target'].nunique()
TFdmde_target_name = TF_dmde_grn.groupby('source')['target'].unique()
TFdmde_target
Out[71]:
source
ELF3      1789
ELK3      1576
EZH2       435
FOXP1     1112
GATA3     2424
LMO2      1177
MSX2       762
OVOL2     2280
PRDM14     485
PRRX2      432
RARB      1187
TCF7       684
ZIC5      2315
Name: target, dtype: int64
In [72]:
TFasTargets = TF_dmde_grn[TF_dmde_grn['target'].isin(TF_GRNiPSC)]
TFdmde_TF = TFasTargets.groupby('source')['target'].nunique()
TFdmde_TF
Out[72]:
source
ELF3      149
ELK3      125
EZH2       42
FOXP1      86
GATA3     178
LMO2       93
MSX2       65
OVOL2     184
PRDM14     40
PRRX2      47
RARB       97
TCF7       64
ZIC5      187
Name: target, dtype: int64
In [73]:
TFdmde_TF_list = TFasTargets.groupby('source')['target'].unique()
TFdmde_TF_list
Out[73]:
source
ELF3      [AFP, AIRE, ALX3, ARID5B, ATF3, ATF5, ATOH1, A...
ELK3      [AFP, AIRE, ARID5B, ATF3, ATOH1, ATOH8, BCL6, ...
EZH2      [ATF3, BCL6, CREB3L1, DDIT3, DLX1, E2F7, FEZF1...
FOXP1     [ARID5B, ATF3, ATOH8, BCL6, CREB3L1, CREB3L3, ...
GATA3     [AFP, AIRE, ALX3, ARID5B, ATF3, ATOH1, ATOH8, ...
LMO2      [AIRE, ATF3, ATOH8, BCL3, CREB3L1, CREB3L3, CT...
MSX2      [AFP, ALX3, ARID5B, ATF3, ATOH8, BCL6, BHLHE40...
OVOL2     [AIRE, ALX3, ARNT2, ATF3, ATOH1, ATOH8, BCL3, ...
PRDM14    [ALX3, ATF3, ATOH8, EBF3, EGR1, FOSL1, FOSL2, ...
PRRX2     [ALX3, ARID5B, ATOH8, BCL6, CREB3L1, DLX1, DLX...
RARB      [ARNT2, ATF3, ATOH8, BCL3, BCL6, BHLHE40, CREB...
TCF7      [AIRE, ATF3, CREB3L1, DDIT3, EGR1, EGR2, ELF3,...
ZIC5      [AIRE, ALX3, ARNT2, ATF3, ATOH1, ATOH8, BCL3, ...
Name: target, dtype: object
In [74]:
ExpectedasTargets = TF_dmde_grn[TF_dmde_grn['target'].isin(expected_all)]
TFdmde_expected = ExpectedasTargets.groupby('source')['target'].nunique()
TFdmde_expected
Out[74]:
source
ELF3      105
ELK3       80
EZH2       22
FOXP1      75
GATA3     148
LMO2       80
MSX2       42
OVOL2     131
PRDM14     26
PRRX2      23
RARB       77
TCF7       44
ZIC5      132
Name: target, dtype: int64
In [75]:
TFdmde_expected_list = ExpectedasTargets.groupby('source')['target'].unique()
TFdmde_expected_list
Out[75]:
source
ELF3      [A4GALT, ACADL, ADGRD1, ANK1, ANO2, APLNR, ARH...
ELK3      [ANK1, ANO2, APLNR, ARHGAP8, ATRNL1, BCAR3, BI...
EZH2      [ANK1, ANO2, B3GNT2, ENPP2, EPHB3, HSPA8, ITM2...
FOXP1     [ANO2, BCAR3, BIN1, BMPR1A, CAPN2, CDO1, COL17...
GATA3     [A4GALT, ABCA4, ACADL, ADGRD1, ANK1, ANO2, AQP...
LMO2      [A4GALT, ANK1, ANO2, ATRNL1, AUTS2, B3GNT2, BM...
MSX2      [CDO1, ELK3, EZH2, FAM174B, FN1, FNDC4, FOXP1,...
OVOL2     [A4GALT, ACADL, ADGRD1, ANK1, ANO2, ARHGAP8, A...
PRDM14    [ANO2, BCAR3, DRAXIN, FN1, FXYD3, GPRC5B, IL6R...
PRRX2     [ANO2, CDO1, COL17A1, EFEMP1, ELK3, EPHB3, EZH...
RARB      [ABCA4, ANO2, APLNR, AXIN2, B3GNT2, BCAR3, BIN...
TCF7      [ADGRD1, ATRNL1, AXIN2, B3GNT2, CDO1, CSF3R, E...
ZIC5      [A4GALT, ACADL, ANK1, ANO2, ARHGAP8, ATRNL1, A...
Name: target, dtype: object
In [76]:
all_targets = set()
for targets in TFdmde_target_name:
    all_targets.update(targets)

all_targets = list(all_targets)

presence_matrix = pd.DataFrame(0, index=all_targets, columns=TFdmde_target_name.index)

for tf in TFdmde_target_name.index:
    presence_matrix[tf] = presence_matrix.index.isin(TFdmde_target_name[tf]).astype(int)

intersection_matrix = pd.DataFrame(np.zeros((len(TFdmde_target_name.index), len(TFdmde_target_name.index))), index=TFdmde_target_name.index, columns=TFdmde_target_name.index)

for tf1, tf2 in combinations(TFdmde_target_name.index, 2):
    common_targets = (presence_matrix[tf1] & presence_matrix[tf2]).sum()
    intersection_matrix.loc[tf1, tf2] = common_targets
    intersection_matrix.loc[tf2, tf1] = common_targets

mask = np.tril(np.ones_like(intersection_matrix, dtype=bool), k=-1)

plt.figure(figsize=(8, 6))
sns.heatmap(intersection_matrix, annot=True, cmap='viridis', mask=mask, fmt='g', linewidths=.5)
plt.show()
In [77]:
target_shares = presence_matrix.sum(axis=1)

filtered_target_shares = target_shares[target_shares >= 1]

target_count_by_tf = filtered_target_shares.value_counts().sort_index()

plt.figure(figsize=(12, 6))
sns.barplot(x=target_count_by_tf.index, y=target_count_by_tf.values, palette='viridis')
plt.xlabel('Number of TFs Sharing the same Target')
plt.ylabel('Number of Targets')
plt.xticks(rotation=0)
plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.show()

/tmp/ipykernel_3576080/2558451435.py:8: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

  sns.barplot(x=target_count_by_tf.index, y=target_count_by_tf.values, palette='viridis')
In [78]:
common_targets = presence_matrix[presence_matrix.sum(axis=1) == len(TFdmde_target_name.index)]

shared_targets = common_targets.index.tolist()

print("Targets shared by all 13 TFs:")
print(shared_targets)

Targets shared by all 13 TFs:
['ZFHX2', 'DCAF11', 'PLA2G3', 'ENO2', 'RNF220', 'YPEL3', 'FOXP4', 'TFEB', 'PCDHB5', 'OTX2', 'PHLDB1', 'ACIN1']

Enrichment 13 TFs¶

In [79]:
DE_universe = pd.read_excel("edgeR_results.xlsx")
In [80]:
DE_universe
Out[80]:
Unnamed: 0 logFC.celltypes_leveledhEGCLC logFC.celltypes_leveledhPGCLC logFC.celltypes_levelediMeLC logCPM LR PValue FDR
0 AHNAK 12.040376 16.812560 12.702891 6.819067 2650.513617 0.000000 0.000000
1 TCL1A 11.390463 17.657546 0.000000 7.638238 3871.022212 0.000000 0.000000
2 ARID5B 11.237896 15.711110 0.000000 5.602953 2040.849486 0.000000 0.000000
3 DLL1 -11.212241 3.561523 -11.212241 4.571846 1551.104568 0.000000 0.000000
4 ALOX5 -11.162451 4.423817 -11.162451 5.437250 2265.240747 0.000000 0.000000
... ... ... ... ... ... ... ... ...
12170 NSFL1C 0.027774 -0.006785 -0.001546 7.090194 0.079787 0.994147 0.994474
12171 INTS4 -0.023568 0.002614 0.019111 6.145675 0.056055 0.996529 0.996775
12172 COA3 0.006039 0.030217 0.015463 6.922481 0.050343 0.997041 0.997205
12173 HPF1 -0.030257 -0.002054 -0.034149 4.989105 0.045044 0.997492 0.997573
12174 AAMP 0.005527 -0.006158 0.032127 8.277898 0.039304 0.997952 0.997952

12175 rows × 8 columns

In [81]:
DE_universe_genes = DE_universe['Unnamed: 0'].unique().tolist()
len(DE_universe_genes)
Out[81]:
12175
In [82]:
DM_universe = pyreadr.read_r('GeneUniverse.rds')
In [83]:
DM_universe
Out[83]:
OrderedDict([(None,
                    ensembl_gene_id_version  ensembl_gene_id  hgnc_symbol  \
              0           ENSG00000223972.5  ENSG00000223972      DDX11L1   
              1           ENSG00000278267.1  ENSG00000278267    MIR6859-1   
              2           ENSG00000243485.5  ENSG00000243485  MIR1302-2HG   
              3           ENSG00000237613.2  ENSG00000237613      FAM138A   
              4           ENSG00000186092.6  ENSG00000186092        OR4F5   
              ...                       ...              ...          ...   
              36011       ENSG00000188399.5  ENSG00000188399    ANKRD36P1   
              36012       ENSG00000215506.5  ENSG00000215506      TPTE2P4   
              36013       ENSG00000231514.1  ENSG00000231514       CCNQP2   
              36014       ENSG00000237917.1  ENSG00000237917      PARP4P1   
              36015       ENSG00000235857.1  ENSG00000235857      CTBP2P1   
              
                                           gene_biotype entrez_gene_id  
              0      transcribed_unprocessed_pseudogene      100287102  
              1                                   miRNA      102466751  
              2                                  lncRNA      107985730  
              3                                  lncRNA         645520  
              4                          protein_coding          79501  
              ...                                   ...            ...  
              36011              unprocessed_pseudogene      100132420  
              36012              unprocessed_pseudogene         286573  
              36013                processed_pseudogene      100421487  
              36014              unprocessed_pseudogene         347613  
              36015                processed_pseudogene         352905  
              
              [36016 rows x 5 columns])])
In [84]:
DM_universe = OrderedDict(DM_universe)
In [85]:
DM_universe = list(DM_universe.values())[0]
DM_universe
Out[85]:
ensembl_gene_id_version ensembl_gene_id hgnc_symbol gene_biotype entrez_gene_id
0 ENSG00000223972.5 ENSG00000223972 DDX11L1 transcribed_unprocessed_pseudogene 100287102
1 ENSG00000278267.1 ENSG00000278267 MIR6859-1 miRNA 102466751
2 ENSG00000243485.5 ENSG00000243485 MIR1302-2HG lncRNA 107985730
3 ENSG00000237613.2 ENSG00000237613 FAM138A lncRNA 645520
4 ENSG00000186092.6 ENSG00000186092 OR4F5 protein_coding 79501
... ... ... ... ... ...
36011 ENSG00000188399.5 ENSG00000188399 ANKRD36P1 unprocessed_pseudogene 100132420
36012 ENSG00000215506.5 ENSG00000215506 TPTE2P4 unprocessed_pseudogene 286573
36013 ENSG00000231514.1 ENSG00000231514 CCNQP2 processed_pseudogene 100421487
36014 ENSG00000237917.1 ENSG00000237917 PARP4P1 unprocessed_pseudogene 347613
36015 ENSG00000235857.1 ENSG00000235857 CTBP2P1 processed_pseudogene 352905

36016 rows × 5 columns

In [86]:
DM_universe_genes = DM_universe['hgnc_symbol']
In [87]:
DM_universe_genes = list(DM_universe_genes)
len(DM_universe_genes)
Out[87]:
36016
In [88]:
target_GRNiPSC = list(target_GRNiPSC)
TF_GRNiPSC = list(TF_GRNiPSC)
In [89]:
GRN_universo = set(target_GRNiPSC + TF_GRNiPSC)
len(GRN_universo)
Out[89]:
2929
In [90]:
universo_13TFs = set(DE_universe_genes) & set(DM_universe_genes) & set(GRN_universo)
len(universo_13TFs)
Out[90]:
2370
In [91]:
TFDMDE_enrichment_iPSC = compute_enrichment(commonTF_expected_all_GRNiPSC, GRN_iPSC_filt, universo, expected_all, p_value_threshold=0.05)
print(TFDMDE_enrichment_iPSC.sort_values(by='enrichment', ascending=False))

TFDMDE_enrichment_iMeLC = compute_enrichment(commonTF_expected_all_GRNiPSC, GRN_iMeLC_filt, universo, expected_all, p_value_threshold=0.05)
print(TFDMDE_enrichment_iMeLC.sort_values(by='enrichment', ascending=False))

TFDMDE_enrichment_PG = compute_enrichment(commonTF_expected_all_GRNiPSC, GRN_PGCLC_filt, universo, expected_all, p_value_threshold=0.05)
print(TFDMDE_enrichment_PG.sort_values(by='enrichment', ascending=False))

TFDMDE_enrichment_EG = compute_enrichment(commonTF_expected_all_GRNiPSC, GRN_EGCLC_filt, universo, expected_all, p_value_threshold=0.05)
print(TFDMDE_enrichment_EG.sort_values(by='enrichment', ascending=False))

       TF  targetTF  targetDMDE    pValue  enrichment
10   LMO2       243          27  0.000172    3.392257
0   FOXP1       457          42  0.000239    2.908042
9   OVOL2       766          52  0.031013    2.321220
6    ZIC5       874          61  0.011482    2.163107
       TF  targetTF  targetDMDE    pValue  enrichment
0   FOXP1       394          35  0.001473    2.837619
1    TCF7       222          18  0.044615    2.667869
2    RARB       163          15  0.024517    2.610758
10   LMO2       237          20  0.024541    2.598837
9   OVOL2       672          47  0.025025    2.207082
6    ZIC5       851          56  0.043164    1.963779
       TF  targetTF  targetDMDE        pValue  enrichment
0   FOXP1       158          31  1.078574e-10    5.026639
4   PRRX2        61           7  4.020857e-02    4.918540
9   OVOL2       370          59  3.788240e-15    4.145626
8    ELF3       196          34  4.298830e-10    4.031446
12   ELK3       268          33  3.612586e-06    3.704496
10   LMO2       166          20  4.216796e-04    3.683788
2    RARB        91          12  2.774809e-03    3.502433
5    MSX2       106          17  3.337045e-05    3.131220
11  GATA3       183          19  3.422574e-03    3.029065
6    ZIC5       256          30  2.712421e-05    2.860027
       TF  targetTF  targetDMDE    pValue  enrichment
10   LMO2       336          30  0.002957    2.736275
0   FOXP1       495          38  0.011253    2.387144
9   OVOL2       719          50  0.023151    2.205990
In [92]:
plot_enrichment(TFDMDE_enrichment_iPSC, enrichment_threshold=1)
plot_enrichment(TFDMDE_enrichment_iMeLC, enrichment_threshold=1)
plot_enrichment(TFDMDE_enrichment_PG, enrichment_threshold=1)
plot_enrichment(TFDMDE_enrichment_EG, enrichment_threshold=1)
In [93]:
TFDMDE_enrichment_iPSC['celltype'] = 'iPSC'
TFDMDE_enrichment_iMeLC['celltype'] = 'iMeLC'
TFDMDE_enrichment_PG['celltype'] = 'hPGCLC'
TFDMDE_enrichment_EG['celltype'] = 'hEGCLC'

TFDMDE_enrichment_iPSC['-log10(pValue)'] = -np.log10(TFDMDE_enrichment_iPSC['pValue'].replace(0, np.nan))
TFDMDE_enrichment_iMeLC['-log10(pValue)'] = -np.log10(TFDMDE_enrichment_iMeLC['pValue'].replace(0, np.nan))
TFDMDE_enrichment_PG['-log10(pValue)'] = -np.log10(TFDMDE_enrichment_PG['pValue'].replace(0, np.nan))
TFDMDE_enrichment_EG['-log10(pValue)'] = -np.log10(TFDMDE_enrichment_EG['pValue'].replace(0, np.nan))
In [94]:
TFDMDE_enrichment_combined = pd.concat([TFDMDE_enrichment_iPSC, TFDMDE_enrichment_iMeLC, TFDMDE_enrichment_PG, TFDMDE_enrichment_EG], ignore_index=True)
TFDMDE_enrichment_combined
Out[94]:
TF targetTF targetDMDE pValue enrichment celltype -log10(pValue)
0 FOXP1 457 42 2.386655e-04 2.908042 iPSC 3.622210
1 ZIC5 874 61 1.148175e-02 2.163107 iPSC 1.939992
2 OVOL2 766 52 3.101336e-02 2.321220 iPSC 1.508451
3 LMO2 243 27 1.715970e-04 3.392257 iPSC 3.765490
4 FOXP1 394 35 1.472965e-03 2.837619 iMeLC 2.831807
5 TCF7 222 18 4.461530e-02 2.667869 iMeLC 1.350516
6 RARB 163 15 2.451724e-02 2.610758 iMeLC 1.610528
7 ZIC5 851 56 4.316402e-02 1.963779 iMeLC 1.364878
8 OVOL2 672 47 2.502507e-02 2.207082 iMeLC 1.601625
9 LMO2 237 20 2.454149e-02 2.598837 iMeLC 1.610099
10 FOXP1 158 31 1.078574e-10 5.026639 hPGCLC 9.967150
11 RARB 91 12 2.774809e-03 3.502433 hPGCLC 2.556767
12 PRRX2 61 7 4.020857e-02 4.918540 hPGCLC 1.395681
13 MSX2 106 17 3.337045e-05 3.131220 hPGCLC 4.476638
14 ZIC5 256 30 2.712421e-05 2.860027 hPGCLC 4.566643
15 ELF3 196 34 4.298830e-10 4.031446 hPGCLC 9.366650
16 OVOL2 370 59 3.788240e-15 4.145626 hPGCLC 14.421563
17 LMO2 166 20 4.216796e-04 3.683788 hPGCLC 3.375017
18 GATA3 183 19 3.422574e-03 3.029065 hPGCLC 2.465647
19 ELK3 268 33 3.612586e-06 3.704496 hPGCLC 5.442182
20 FOXP1 495 38 1.125316e-02 2.387144 hEGCLC 1.948725
21 OVOL2 719 50 2.315096e-02 2.205990 hEGCLC 1.635431
22 LMO2 336 30 2.956952e-03 2.736275 hEGCLC 2.529156
In [95]:
pivot_table = TFDMDE_enrichment_combined.pivot("celltype", "TF", "enrichment")

plt.figure(figsize=(13, 5))
heatmap = sns.heatmap(
    pivot_table,
    annot=True,
    cmap="Blues",
    fmt="g",
    linewidths=0.5,
    linecolor='gray',
    cbar_kws={'label': 'enrichment'}
)


heatmap.set_xlabel('TFs', size=12)
heatmap.set_ylabel('CellTypes', size=12)

plt.xticks(rotation=45, ha='right')

plt.tight_layout()
plt.show()

/tmp/ipykernel_3576080/1182190980.py:1: FutureWarning: In a future version of pandas all arguments of DataFrame.pivot will be keyword-only.
  pivot_table = TFDMDE_enrichment_combined.pivot("celltype", "TF", "enrichment")
In [123]:
plt.figure(figsize=(12, 4))
sns.set_theme(style="whitegrid")

original_cmap = plt.get_cmap('Blues')
darken_cmap = LinearSegmentedColormap.from_list('dark_Blues', [original_cmap(0.15), original_cmap(1.0)], N=256)

sns.scatterplot(
    data=TFDMDE_enrichment_combined,
    x='TF',
    y='celltype',
    hue='enrichment',
    size='-log10(pValue)',
    sizes=(50, 400),
    palette=darken_cmap
    )

plt.xlabel('TFs', size=11)
plt.ylabel('CellTypes', size=11)
plt.xticks(rotation=45, ha='right')
plt.legend( fontsize='small', bbox_to_anchor=(1.05, 1), loc='upper left')
plt.tight_layout()
plt.show()
In [97]:
GRN_iPSC_13TF = GRN_iPSC_filt[GRN_iPSC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
print(GRN_iPSC_13TF.sort_values(by='coef_abs', ascending=False))

GRN_iPSC_13TF.to_csv('GRN_iPSC_13TF.csv', index=False)

GRN_iMeLC_13TF = GRN_iMeLC_filt[GRN_iMeLC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
print(GRN_iMeLC_13TF.sort_values(by='coef_abs', ascending=False))

GRN_iMeLC_13TF.to_csv('GRN_iMeLC_13TF.csv', index=False)

GRN_hPGCLC_13TF = GRN_PGCLC_filt[GRN_PGCLC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
print(GRN_hPGCLC_13TF.sort_values(by='coef_abs', ascending=False))

GRN_hPGCLC_13TF.to_csv('GRN_hPGCLC_13TF.csv', index=False)

GRN_hEGCLC_13TF = GRN_EGCLC_filt[GRN_EGCLC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
print(GRN_hEGCLC_13TF.sort_values(by='coef_abs', ascending=False))

GRN_hEGCLC_13TF.to_csv('GRN_hEGCLC_13TF.csv', index=False)

        Unnamed: 0 source     target  coef_mean  coef_abs             p  \
248835      248835   ELK3      RACK1  -0.461227  0.461227  1.120887e-18   
134504      134504   ELK3   HIST1H1C  -0.387532  0.387532  4.122939e-10   
135846      135846   ELK3  HIST1H2BH  -0.347674  0.347674  4.379518e-13   
143669      143669   ELK3    HSP90B1   0.295289  0.295289  5.417288e-15   
134452      134452   ZIC5   HIST1H1B  -0.268723  0.268723  1.318423e-08   
...            ...    ...        ...        ...       ...           ...   
318089      318089   ELF3     TRIM36   0.010004  0.010004  3.446703e-09   
215747      215747   ZIC5       PAK4   0.010004  0.010004  7.471403e-07   
216445      216445   ELK3      PAQR5  -0.010002  0.010002  2.691006e-08   
317826      317826   ZIC5     TRIM23  -0.010000  0.010000  2.219330e-10   
338789      338789  GATA3      YWHAB   0.010000  0.010000  2.352963e-09   

            -logp  
248835  17.950438  
134504   9.384793  
135846  12.358574  
143669  14.266218  
134452   7.879945  
...           ...  
318089   8.462596  
215747   6.126598  
216445   7.570085  
317826   9.653778  
338789   8.628385  

[4967 rows x 7 columns]
        Unnamed: 0  source   target  coef_mean  coef_abs             p  \
283997      283997   GATA3    SOAT1   0.712898  0.712898  4.231232e-20   
143669      143669    ELK3  HSP90B1   0.173806  0.173806  5.088399e-12   
286690      286690    ELK3    SPARC   0.148440  0.148440  2.457314e-15   
295161      295161    ELK3    SURF4   0.148018  0.148018  1.801832e-15   
248835      248835    ELK3    RACK1  -0.146893  0.146893  4.015747e-11   
...            ...     ...      ...        ...       ...           ...   
284614      284614   GATA3    SOGA1   0.010007  0.010007  6.960855e-13   
149361      149361    LMO2       IK   0.010004  0.010004  7.753705e-09   
102960      102960    ELK3     FDPS   0.010004  0.010004  2.300454e-08   
292388      292388    EZH2    STAT2  -0.010002  0.010002  1.377573e-10   
83316        83316  PRDM14     EBF3   0.010001  0.010001  3.761337e-11   

            -logp  
283997  19.373533  
143669  11.293419  
286690  14.609539  
295161  14.744286  
248835  10.396234  
...           ...  
284614  12.157337  
149361   8.110491  
102960   7.638187  
292388   9.860885  
83316   10.424658  

[4227 rows x 7 columns]
        Unnamed: 0 source  target  coef_mean  coef_abs             p  \
148643      148643   ELF3    IGF1   0.117080  0.117080  1.045044e-12   
283997      283997  GATA3   SOAT1   0.101784  0.101784  1.376599e-16   
163071      163071   ELK3  KLHL18   0.080774  0.080774  4.521049e-12   
247422      247422   MSX2   RAB15   0.075933  0.075933  8.819942e-16   
72210        72210  OVOL2  CYP1B1   0.075118  0.075118  2.791718e-15   
...            ...    ...     ...        ...       ...           ...   
325181      325181   ELF3     UBC   0.010014  0.010014  2.983674e-15   
108096      108096   RARB     FOS  -0.010009  0.010009  3.999690e-05   
70275        70275   ZIC5    CTSL  -0.010009  0.010009  5.823589e-12   
314410      314410  GATA3    TOP1   0.010004  0.010004  3.731211e-10   
204922      204922   TCF7   NOP56   0.010004  0.010004  9.704790e-14   

            -logp  
148643  11.980865  
283997  15.861192  
163071  11.344761  
247422  15.054534  
72210   14.554128  
...           ...  
325181  14.525249  
108096   4.397974  
70275   11.234809  
314410   9.428150  
204922  13.013014  

[1934 rows x 7 columns]
        Unnamed: 0 source    target  coef_mean  coef_abs             p  \
248835      248835   ELK3     RACK1  -0.555234  0.555234  9.021584e-20   
221513      221513  GATA3      PDPN   0.425710  0.425710  3.703445e-14   
134626      134626   ZIC5  HIST1H1C  -0.343693  0.343693  1.690755e-09   
143669      143669   ELK3   HSP90B1   0.342187  0.342187  3.559395e-13   
134452      134452   ZIC5  HIST1H1B  -0.283132  0.283132  5.070790e-10   
...            ...    ...       ...        ...       ...           ...   
85326        85326   ZIC5     EEF1G   0.010004  0.010004  1.540312e-03   
43555        43555  OVOL2     CCND1  -0.010003  0.010003  1.051781e-06   
293577      293577  OVOL2     STON1   0.010003  0.010003  1.567192e-07   
164849      164849   ELK3   L3MBTL2  -0.010000  0.010000  5.955790e-10   
7326          7326   ELK3  ADAMTSL1   0.010000  0.010000  2.601280e-10   

            -logp  
248835  19.044717  
221513  13.431394  
134626   8.771919  
143669  12.448624  
134452   9.294924  
...           ...  
85326    2.812391  
43555    5.978075  
293577   6.804878  
164849   9.225061  
7326     9.584813  

[5310 rows x 7 columns]
In [98]:
GRN_iPSC_13TF_expectedall = GRN_iPSC_13TF[GRN_iPSC_13TF['target'].isin(expected_all)]
GRN_iPSC_13TF_expectedall.to_csv('GRN_iPSC_13TF_expectedall.csv', index=False)

GRN_iMeLC_13TF_expectedall = GRN_iMeLC_13TF[GRN_iMeLC_13TF['target'].isin(expected_all)]
GRN_iMeLC_13TF_expectedall.to_csv('GRN_iMeLC_13TF_expectedall.csv', index=False)

GRN_hPGCLC_13TF_expectedall = GRN_hPGCLC_13TF[GRN_hPGCLC_13TF['target'].isin(expected_all)]
GRN_hPGCLC_13TF_expectedall.to_csv('GRN_hPGCLC_13TF_expectedall.csv', index=False)

GRN_hEGCLC_13TF_expectedall = GRN_hEGCLC_13TF[GRN_hEGCLC_13TF['target'].isin(expected_all)]
GRN_hEGCLC_13TF_expectedall.to_csv('GRN_hEGCLC_13TF_expectedall.csv', index=False)

Cytoscape¶

hPGCLC¶
In [99]:
GRN_PGCLC_filt
Out[99]:
Unnamed: 0 source target coef_mean coef_abs p -logp
4 4 E2F1 A4GALT -0.011982 0.011982 1.966049e-09 8.706406
6 6 RFX3 A4GALT -0.012635 0.012635 7.385996e-09 8.131591
9 9 KLF4 A4GALT 0.027695 0.027695 1.407787e-13 12.851463
11 11 SP8 A4GALT -0.010359 0.010359 1.637664e-10 9.785775
14 14 FOXB2 A4GALT 0.010374 0.010374 1.412924e-10 9.849881
... ... ... ... ... ... ... ...
350450 350450 REST ZYG11A 0.016032 0.016032 3.348030e-18 17.475211
350453 350453 MEF2A ZYG11A 0.010208 0.010208 9.384473e-14 13.027590
350458 350458 NPAS1 ZYG11A 0.011459 0.011459 6.519771e-13 12.185768
350463 350463 TFAP2C ZYG11A 0.020360 0.020360 2.841033e-19 18.546524
350474 350474 ZNF816 ZYG11A 0.010861 0.010861 1.269887e-14 13.896235

28207 rows × 7 columns

In [100]:
GRN_PGCLC_cyto1 = GRN_PGCLC_filt[GRN_PGCLC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
GRN_PGCLC_cyto1
Out[100]:
Unnamed: 0 source target coef_mean coef_abs p -logp
53 53 LMO2 A4GALT 0.023551 0.023551 1.056805e-15 14.976005
56 56 OVOL2 A4GALT 0.012714 0.012714 6.383285e-12 11.194956
105 105 GATA3 A4GALT -0.017881 0.017881 1.668532e-13 12.777666
414 414 ZIC5 AASS 0.027348 0.027348 1.612248e-22 21.792568
440 440 MSX2 AASS -0.031137 0.031137 8.849539e-20 19.053079
... ... ... ... ... ... ... ...
349451 349451 ZIC5 ZNF850 0.013029 0.013029 7.244267e-12 11.140006
349547 349547 ELF3 ZPR1 0.011409 0.011409 1.611430e-13 12.792789
349556 349556 OVOL2 ZPR1 0.012896 0.012896 2.215100e-10 9.654607
350277 350277 ELK3 ZSWIM7 0.014401 0.014401 3.712457e-17 16.430339
350427 350427 OVOL2 ZYG11A 0.012407 0.012407 7.396718e-17 16.130961

1934 rows × 7 columns

In [101]:
GRN_PGCLC_cyto2 = GRN_PGCLC_filt[GRN_PGCLC_filt['target'].isin(commonTF_expected_all_GRNiPSC)]
GRN_PGCLC_cyto2
Out[101]:
Unnamed: 0 source target coef_mean coef_abs p -logp
88695 88695 OVOL1 ELF3 0.010808 0.010808 7.244517e-15 14.139991
88697 88697 MAF ELF3 0.011553 0.011553 8.798807e-14 13.055576
88965 88965 MAF ELK3 0.011011 0.011011 1.717420e-17 16.765123
88974 88974 ETV5 ELK3 0.011128 0.011128 8.653431e-16 15.062812
88994 88994 BCL6 ELK3 -0.012313 0.012313 8.308876e-13 12.080458
... ... ... ... ... ... ... ...
342407 342407 ARID5B ZIC5 -0.011111 0.011111 6.059643e-18 17.217553
342410 342410 ZIC2 ZIC5 0.013522 0.013522 2.449429e-17 16.610935
342433 342433 TFAP2C ZIC5 -0.012293 0.012293 2.368252e-21 20.625572
342439 342439 MEF2C ZIC5 0.012175 0.012175 4.485344e-13 12.348204
342447 342447 MSX2 ZIC5 -0.012782 0.012782 3.523284e-15 14.453052

104 rows × 7 columns

In [102]:
GRN_PGCLC_cyto3 =  pd.concat([GRN_PGCLC_cyto1, GRN_PGCLC_cyto2])
GRN_PGCLC_cyto4 = GRN_PGCLC_cyto3.drop_duplicates()
GRN_PGCLC_cyto4
Out[102]:
Unnamed: 0 source target coef_mean coef_abs p -logp
53 53 LMO2 A4GALT 0.023551 0.023551 1.056805e-15 14.976005
56 56 OVOL2 A4GALT 0.012714 0.012714 6.383285e-12 11.194956
105 105 GATA3 A4GALT -0.017881 0.017881 1.668532e-13 12.777666
414 414 ZIC5 AASS 0.027348 0.027348 1.612248e-22 21.792568
440 440 MSX2 AASS -0.031137 0.031137 8.849539e-20 19.053079
... ... ... ... ... ... ... ...
342398 342398 POU3F4 ZIC5 0.010349 0.010349 2.434231e-12 11.613638
342407 342407 ARID5B ZIC5 -0.011111 0.011111 6.059643e-18 17.217553
342410 342410 ZIC2 ZIC5 0.013522 0.013522 2.449429e-17 16.610935
342433 342433 TFAP2C ZIC5 -0.012293 0.012293 2.368252e-21 20.625572
342439 342439 MEF2C ZIC5 0.012175 0.012175 4.485344e-13 12.348204

2028 rows × 7 columns

In [103]:
GRN_PGCLC_cyto_final = GRN_PGCLC_cyto4[GRN_PGCLC_cyto4['target'].isin(expected_all)]
GRN_PGCLC_cyto_final
Out[103]:
Unnamed: 0 source target coef_mean coef_abs p -logp
53 53 LMO2 A4GALT 0.023551 0.023551 1.056805e-15 14.976005
56 56 OVOL2 A4GALT 0.012714 0.012714 6.383285e-12 11.194956
105 105 GATA3 A4GALT -0.017881 0.017881 1.668532e-13 12.777666
2481 2481 ELF3 ACADL 0.018064 0.018064 1.473265e-11 10.831719
2488 2488 OVOL2 ACADL 0.011669 0.011669 4.484103e-09 8.348324
... ... ... ... ... ... ... ...
342398 342398 POU3F4 ZIC5 0.010349 0.010349 2.434231e-12 11.613638
342407 342407 ARID5B ZIC5 -0.011111 0.011111 6.059643e-18 17.217553
342410 342410 ZIC2 ZIC5 0.013522 0.013522 2.449429e-17 16.610935
342433 342433 TFAP2C ZIC5 -0.012293 0.012293 2.368252e-21 20.625572
342439 342439 MEF2C ZIC5 0.012175 0.012175 4.485344e-13 12.348204

363 rows × 7 columns

In [104]:
GRN_PGCLC_cyto_final.to_csv('GRN_PGC_cytoscape.csv', index=False)
hiPSC¶
In [105]:
GRN_iPSC_filt
Out[105]:
Unnamed: 0 source target coef_mean coef_abs p -logp
13 13 EGR1 A4GALT -0.011337 0.011337 2.955064e-11 10.529433
21 21 FOS A4GALT 0.011398 0.011398 1.493396e-09 8.825825
24 24 FOXP4 A4GALT 0.023915 0.023915 4.643730e-14 13.333133
25 25 SNAI1 A4GALT 0.010837 0.010837 2.070093e-13 12.684010
29 29 ELF3 A4GALT 0.013303 0.013303 5.879207e-13 12.230681
... ... ... ... ... ... ... ...
350434 350434 ZNF165 ZYG11A 0.013430 0.013430 9.960155e-12 11.001734
350451 350451 BHLHE40 ZYG11A 0.011117 0.011117 1.016146e-08 7.993044
350461 350461 ATF3 ZYG11A 0.014585 0.014585 5.512019e-15 14.258689
350462 350462 ZIC3 ZYG11A 0.022429 0.022429 7.037211e-19 18.152599
350472 350472 BCL3 ZYG11A 0.016571 0.016571 5.683222e-15 14.245405

77471 rows × 7 columns

In [106]:
GRN_iPSC_cyto1 = GRN_iPSC_filt[GRN_iPSC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
GRN_iPSC_cyto1
Out[106]:
Unnamed: 0 source target coef_mean coef_abs p -logp
29 29 ELF3 A4GALT 0.013303 0.013303 5.879207e-13 12.230681
53 53 LMO2 A4GALT 0.017735 0.017735 1.102182e-13 12.957747
90 90 ZIC5 A4GALT 0.013395 0.013395 1.373494e-07 6.862173
154 154 ELF3 AARSD1 0.018138 0.018138 1.601966e-10 9.795347
155 155 RARB AARSD1 0.013204 0.013204 2.673889e-12 11.572857
... ... ... ... ... ... ... ...
349593 349593 GATA3 ZPR1 -0.027981 0.027981 4.769950e-14 13.321486
349804 349804 ELK3 ZSCAN2 0.030781 0.030781 8.699941e-12 11.060484
349829 349829 ELF3 ZSCAN2 -0.017729 0.017729 2.283838e-07 6.641335
350277 350277 ELK3 ZSWIM7 -0.014034 0.014034 1.158973e-06 5.935927
350356 350356 ZIC5 ZSWIM7 -0.025648 0.025648 4.506691e-12 11.346142

4967 rows × 7 columns

In [107]:
GRN_iPSC_cyto2 = GRN_iPSC_filt[GRN_iPSC_filt['target'].isin(commonTF_expected_all_GRNiPSC)]
GRN_iPSC_cyto2
Out[107]:
Unnamed: 0 source target coef_mean coef_abs p -logp
88695 88695 OVOL1 ELF3 -0.013847 0.013847 1.358852e-15 14.866828
88708 88708 ELK3 ELF3 0.013378 0.013378 1.916311e-14 13.717534
88709 88709 EGR1 ELF3 0.010768 0.010768 1.250223e-08 7.903012
88710 88710 TFCP2 ELF3 0.011534 0.011534 9.640954e-11 10.015880
88716 88716 KLF5 ELF3 0.016137 0.016137 2.282798e-13 12.641532
... ... ... ... ... ... ... ...
342443 342443 GATA3 ZIC5 0.015403 0.015403 1.242694e-10 9.905636
342444 342444 SOX15 ZIC5 0.010086 0.010086 2.011335e-04 3.696516
342448 342448 SP3 ZIC5 0.058750 0.058750 1.746419e-11 10.757851
342449 342449 ZBTB6 ZIC5 0.010077 0.010077 1.321807e-06 5.878832
342453 342453 ZNF816 ZIC5 -0.011479 0.011479 8.976590e-11 10.046889

326 rows × 7 columns

In [108]:
GRN_iPSC_cyto3 =  pd.concat([GRN_iPSC_cyto1, GRN_iPSC_cyto2])
GRN_iPSC_cyto4 = GRN_iPSC_cyto3.drop_duplicates()
GRN_iPSC_cyto4
Out[108]:
Unnamed: 0 source target coef_mean coef_abs p -logp
29 29 ELF3 A4GALT 0.013303 0.013303 5.879207e-13 12.230681
53 53 LMO2 A4GALT 0.017735 0.017735 1.102182e-13 12.957747
90 90 ZIC5 A4GALT 0.013395 0.013395 1.373494e-07 6.862173
154 154 ELF3 AARSD1 0.018138 0.018138 1.601966e-10 9.795347
155 155 RARB AARSD1 0.013204 0.013204 2.673889e-12 11.572857
... ... ... ... ... ... ... ...
342441 342441 BHLHE41 ZIC5 0.017330 0.017330 2.028516e-07 6.692822
342444 342444 SOX15 ZIC5 0.010086 0.010086 2.011335e-04 3.696516
342448 342448 SP3 ZIC5 0.058750 0.058750 1.746419e-11 10.757851
342449 342449 ZBTB6 ZIC5 0.010077 0.010077 1.321807e-06 5.878832
342453 342453 ZNF816 ZIC5 -0.011479 0.011479 8.976590e-11 10.046889

5268 rows × 7 columns

In [109]:
GRN_iPSC_cyto_final = GRN_iPSC_cyto4[GRN_iPSC_cyto4['target'].isin(expected_all)]
GRN_iPSC_cyto_final
Out[109]:
Unnamed: 0 source target coef_mean coef_abs p -logp
29 29 ELF3 A4GALT 0.013303 0.013303 5.879207e-13 12.230681
53 53 LMO2 A4GALT 0.017735 0.017735 1.102182e-13 12.957747
90 90 ZIC5 A4GALT 0.013395 0.013395 1.373494e-07 6.862173
15061 15061 ELF3 ANK1 0.015547 0.015547 2.388566e-12 11.621863
19322 19322 OVOL2 ARHGAP8 0.033861 0.033861 1.025319e-09 8.989141
... ... ... ... ... ... ... ...
342441 342441 BHLHE41 ZIC5 0.017330 0.017330 2.028516e-07 6.692822
342444 342444 SOX15 ZIC5 0.010086 0.010086 2.011335e-04 3.696516
342448 342448 SP3 ZIC5 0.058750 0.058750 1.746419e-11 10.757851
342449 342449 ZBTB6 ZIC5 0.010077 0.010077 1.321807e-06 5.878832
342453 342453 ZNF816 ZIC5 -0.011479 0.011479 8.976590e-11 10.046889

629 rows × 7 columns

In [110]:
GRN_iPSC_cyto_final.to_csv('GRN_iPSC_cytoscape.csv', index=False)
iMeLC¶
In [111]:
GRN_iMeLC_filt
Out[111]:
Unnamed: 0 source target coef_mean coef_abs p -logp
3 3 MAF A4GALT 0.010437 0.010437 9.697370e-13 12.013346
12 12 TFCP2 A4GALT 0.011424 0.011424 6.991379e-13 12.155437
45 45 POLR3G A4GALT 0.013872 0.013872 1.354756e-11 10.868139
50 50 ZFP57 A4GALT 0.016220 0.016220 2.714614e-15 14.566292
56 56 OVOL2 A4GALT 0.016652 0.016652 6.244700e-14 13.204488
... ... ... ... ... ... ... ...
350434 350434 ZNF165 ZYG11A 0.017363 0.017363 4.585605e-12 11.338603
350441 350441 FOXH1 ZYG11A 0.023062 0.023062 2.598020e-14 13.585357
350450 350450 REST ZYG11A 0.014005 0.014005 1.156650e-14 13.936798
350452 350452 ZIC2 ZYG11A -0.015993 0.015993 2.509253e-11 10.600456
350463 350463 TFAP2C ZYG11A 0.011303 0.011303 1.144868e-13 12.941244

61767 rows × 7 columns

In [112]:
GRN_iMeLC_cyto1 = GRN_iMeLC_filt[GRN_iMeLC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
GRN_iMeLC_cyto1
Out[112]:
Unnamed: 0 source target coef_mean coef_abs p -logp
56 56 OVOL2 A4GALT 0.016652 0.016652 6.244700e-14 13.204488
155 155 RARB AARSD1 0.011641 0.011641 2.733939e-12 11.563211
187 187 LMO2 AARSD1 0.010155 0.010155 1.422782e-12 11.846862
230 230 ZIC5 AARSD1 0.016098 0.016098 2.562564e-11 10.591325
278 278 ELK3 AASS 0.070405 0.070405 1.770125e-16 15.751996
... ... ... ... ... ... ... ...
349829 349829 ELF3 ZSCAN2 0.010415 0.010415 2.157605e-09 8.666028
350277 350277 ELK3 ZSWIM7 -0.024761 0.024761 4.669515e-14 13.330728
350284 350284 PRRX2 ZSWIM7 0.018580 0.018580 8.251371e-14 13.083474
350298 350298 ELF3 ZSWIM7 -0.021717 0.021717 3.916052e-15 14.407152
350319 350319 FOXP1 ZSWIM7 -0.020975 0.020975 3.589840e-13 12.444925

4227 rows × 7 columns

In [113]:
GRN_iMeLC_cyto2 = GRN_iMeLC_filt[GRN_iMeLC_filt['target'].isin(commonTF_expected_all_GRNiPSC)]
GRN_iMeLC_cyto2
Out[113]:
Unnamed: 0 source target coef_mean coef_abs p -logp
88962 88962 TCF7L1 ELK3 -0.011948 0.011948 9.778770e-10 9.009716
88976 88976 SP8 ELK3 -0.010335 0.010335 2.972205e-13 12.526921
88977 88977 TFCP2 ELK3 0.017199 0.017199 2.878845e-11 10.540782
88995 88995 FOXO1 ELK3 0.013059 0.013059 1.139062e-12 11.943452
89013 89013 CREB3L1 ELK3 0.013361 0.013361 1.799579e-12 11.744829
... ... ... ... ... ... ... ...
342436 342436 KLF6 ZIC5 0.010974 0.010974 1.684605e-11 10.773502
342438 342438 PKNOX2 ZIC5 -0.010062 0.010062 9.342705e-10 9.029527
342441 342441 BHLHE41 ZIC5 0.027396 0.027396 3.677205e-13 12.434482
342444 342444 SOX15 ZIC5 -0.010681 0.010681 1.158516e-08 7.936098
342448 342448 SP3 ZIC5 0.032673 0.032673 3.169674e-15 14.498985

267 rows × 7 columns

In [114]:
GRN_iMeLC_cyto3 =  pd.concat([GRN_iMeLC_cyto1, GRN_iMeLC_cyto2])
GRN_iMeLC_cyto4 = GRN_iMeLC_cyto3.drop_duplicates()
GRN_iMeLC_cyto4
Out[114]:
Unnamed: 0 source target coef_mean coef_abs p -logp
56 56 OVOL2 A4GALT 0.016652 0.016652 6.244700e-14 13.204488
155 155 RARB AARSD1 0.011641 0.011641 2.733939e-12 11.563211
187 187 LMO2 AARSD1 0.010155 0.010155 1.422782e-12 11.846862
230 230 ZIC5 AARSD1 0.016098 0.016098 2.562564e-11 10.591325
278 278 ELK3 AASS 0.070405 0.070405 1.770125e-16 15.751996
... ... ... ... ... ... ... ...
342436 342436 KLF6 ZIC5 0.010974 0.010974 1.684605e-11 10.773502
342438 342438 PKNOX2 ZIC5 -0.010062 0.010062 9.342705e-10 9.029527
342441 342441 BHLHE41 ZIC5 0.027396 0.027396 3.677205e-13 12.434482
342444 342444 SOX15 ZIC5 -0.010681 0.010681 1.158516e-08 7.936098
342448 342448 SP3 ZIC5 0.032673 0.032673 3.169674e-15 14.498985

4472 rows × 7 columns

In [115]:
GRN_iMeLC_cyto_final = GRN_iMeLC_cyto4[GRN_iMeLC_cyto4['target'].isin(expected_all)]
GRN_iMeLC_cyto_final
Out[115]:
Unnamed: 0 source target coef_mean coef_abs p -logp
56 56 OVOL2 A4GALT 0.016652 0.016652 6.244700e-14 13.204488
17418 17418 ELK3 APLNR 0.020163 0.020163 1.699266e-12 11.769739
19287 19287 ELK3 ARHGAP8 0.013718 0.013718 8.098435e-07 6.091599
19322 19322 OVOL2 ARHGAP8 0.032714 0.032714 3.410992e-11 10.467119
19347 19347 ZIC5 ARHGAP8 0.015903 0.015903 3.001244e-10 9.522699
... ... ... ... ... ... ... ...
342436 342436 KLF6 ZIC5 0.010974 0.010974 1.684605e-11 10.773502
342438 342438 PKNOX2 ZIC5 -0.010062 0.010062 9.342705e-10 9.029527
342441 342441 BHLHE41 ZIC5 0.027396 0.027396 3.677205e-13 12.434482
342444 342444 SOX15 ZIC5 -0.010681 0.010681 1.158516e-08 7.936098
342448 342448 SP3 ZIC5 0.032673 0.032673 3.169674e-15 14.498985

541 rows × 7 columns

In [116]:
GRN_iMeLC_cyto_final.to_csv('GRN_iMeLC_cytoscape.csv', index=False)
hEGCLC¶
In [117]:
GRN_EGCLC_filt
Out[117]:
Unnamed: 0 source target coef_mean coef_abs p -logp
4 4 E2F1 A4GALT 0.012186 0.012186 2.743329e-10 9.561722
10 10 MYB A4GALT -0.011644 0.011644 1.174329e-06 5.930210
13 13 EGR1 A4GALT -0.021030 0.021030 2.029755e-09 8.692556
15 15 ZNF143 A4GALT -0.010884 0.010884 1.777862e-10 9.750102
33 33 KLF2 A4GALT -0.010645 0.010645 5.090827e-11 10.293212
... ... ... ... ... ... ... ...
350359 350359 NPAS1 ZSWIM7 -0.013866 0.013866 3.799716e-13 12.420249
350363 350363 PBX2 ZSWIM7 0.046887 0.046887 2.019494e-13 12.694757
350367 350367 KLF6 ZSWIM7 0.033844 0.033844 2.047703e-14 13.688733
350415 350415 TCF7L2 ZYG11A 0.010945 0.010945 1.534989e-10 9.813895
350450 350450 REST ZYG11A 0.011057 0.011057 4.454638e-18 17.351188

82575 rows × 7 columns

In [118]:
GRN_EGCLC_cyto1 = GRN_EGCLC_filt[GRN_EGCLC_filt['source'].isin(commonTF_expected_all_GRNiPSC)]
GRN_EGCLC_cyto1
Out[118]:
Unnamed: 0 source target coef_mean coef_abs p -logp
53 53 LMO2 A4GALT 0.018656 0.018656 7.913373e-12 11.101638
129 129 ELK3 AARSD1 0.021767 0.021767 3.068825e-12 11.513028
187 187 LMO2 AARSD1 -0.010638 0.010638 1.295345e-13 12.887615
191 191 OVOL2 AARSD1 0.023981 0.023981 2.207090e-15 14.656180
230 230 ZIC5 AARSD1 -0.026104 0.026104 2.995687e-12 11.523504
... ... ... ... ... ... ... ...
349991 349991 LMO2 ZSCAN25 -0.010526 0.010526 1.772576e-17 16.751395
350277 350277 ELK3 ZSWIM7 -0.046259 0.046259 6.755587e-15 14.170337
350284 350284 PRRX2 ZSWIM7 0.015225 0.015225 2.059283e-11 10.686284
350298 350298 ELF3 ZSWIM7 -0.012967 0.012967 6.320867e-11 10.199223
350356 350356 ZIC5 ZSWIM7 -0.022530 0.022530 4.561422e-08 7.340900

5310 rows × 7 columns

In [119]:
GRN_EGCLC_cyto2 = GRN_EGCLC_filt[GRN_EGCLC_filt['target'].isin(commonTF_expected_all_GRNiPSC)]
GRN_EGCLC_cyto2
Out[119]:
Unnamed: 0 source target coef_mean coef_abs p -logp
88695 88695 OVOL1 ELF3 -0.025000 0.025000 7.044954e-16 15.152122
88704 88704 ID3 ELF3 -0.012820 0.012820 9.211341e-12 11.035677
88710 88710 TFCP2 ELF3 0.012519 0.012519 4.909585e-15 14.308955
88717 88717 ETS2 ELF3 -0.010985 0.010985 3.445869e-13 12.462701
88718 88718 HIC1 ELF3 0.013807 0.013807 1.291168e-12 11.889017
... ... ... ... ... ... ... ...
342432 342432 ZIC3 ZIC5 0.014187 0.014187 1.873290e-08 7.727395
342441 342441 BHLHE41 ZIC5 0.018040 0.018040 2.624662e-12 11.580927
342443 342443 GATA3 ZIC5 0.014054 0.014054 2.158350e-13 12.665878
342448 342448 SP3 ZIC5 0.082432 0.082432 4.051028e-15 14.392435
342449 342449 ZBTB6 ZIC5 0.014288 0.014288 1.470271e-08 7.832603

344 rows × 7 columns

In [120]:
GRN_EGCLC_cyto3 =  pd.concat([GRN_EGCLC_cyto1, GRN_EGCLC_cyto2])
GRN_EGCLC_cyto4 = GRN_EGCLC_cyto3.drop_duplicates()
GRN_EGCLC_cyto4
Out[120]:
Unnamed: 0 source target coef_mean coef_abs p -logp
53 53 LMO2 A4GALT 0.018656 0.018656 7.913373e-12 11.101638
129 129 ELK3 AARSD1 0.021767 0.021767 3.068825e-12 11.513028
187 187 LMO2 AARSD1 -0.010638 0.010638 1.295345e-13 12.887615
191 191 OVOL2 AARSD1 0.023981 0.023981 2.207090e-15 14.656180
230 230 ZIC5 AARSD1 -0.026104 0.026104 2.995687e-12 11.523504
... ... ... ... ... ... ... ...
342426 342426 HAND2 ZIC5 -0.016372 0.016372 1.269484e-11 10.896373
342432 342432 ZIC3 ZIC5 0.014187 0.014187 1.873290e-08 7.727395
342441 342441 BHLHE41 ZIC5 0.018040 0.018040 2.624662e-12 11.580927
342448 342448 SP3 ZIC5 0.082432 0.082432 4.051028e-15 14.392435
342449 342449 ZBTB6 ZIC5 0.014288 0.014288 1.470271e-08 7.832603

5630 rows × 7 columns

In [121]:
GRN_EGCLC_cyto_final = GRN_EGCLC_cyto4[GRN_EGCLC_cyto4['target'].isin(expected_all)]
GRN_EGCLC_cyto_final
Out[121]:
Unnamed: 0 source target coef_mean coef_abs p -logp
53 53 LMO2 A4GALT 0.018656 0.018656 7.913373e-12 11.101638
15061 15061 ELF3 ANK1 -0.015010 0.015010 4.326182e-10 9.363895
15094 15094 OVOL2 ANK1 -0.012912 0.012912 6.356422e-12 11.196787
19287 19287 ELK3 ARHGAP8 -0.024587 0.024587 3.972067e-09 8.400983
19322 19322 OVOL2 ARHGAP8 0.027408 0.027408 1.226135e-07 6.911462
... ... ... ... ... ... ... ...
342426 342426 HAND2 ZIC5 -0.016372 0.016372 1.269484e-11 10.896373
342432 342432 ZIC3 ZIC5 0.014187 0.014187 1.873290e-08 7.727395
342441 342441 BHLHE41 ZIC5 0.018040 0.018040 2.624662e-12 11.580927
342448 342448 SP3 ZIC5 0.082432 0.082432 4.051028e-15 14.392435
342449 342449 ZBTB6 ZIC5 0.014288 0.014288 1.470271e-08 7.832603

658 rows × 7 columns

In [122]:
GRN_EGCLC_cyto_final.to_csv('GRN_EGC_cytoscape.csv', index=False)