Table 3.

Median differences in correlation among p53 pathway genes between non p53-mutated and p53-mutated samples

Gene	Median z-Score Difference	Empirical p-Value	Adjusted p-Value
SERPINB5	-0.730683588	p < 1e-4	p < 1e-4
ATF3	-0.666702829	p < 1e-4	p < 1e-4
FAS	-0.590308897	p < 1e-4	p < 1e-4
ARID3A	0.584112447	p < 1e-4	p < 1e-4
RPS12	-0.565836003	p < 1e-4	p < 1e-4
SPHK1	-0.56379232	p < 1e-4	p < 1e-4
LDHB	-0.562303813	p < 1e-4	p < 1e-4
MET	-0.527886541	p < 1e-4	p < 1e-4
CX3CL1	-0.511204357	p < 1e-4	p < 1e-4
KRT17	-0.504964779	p < 1e-4	p < 1e-4
FOXA1	0.482088394	p < 1e-4	p < 1e-4
IGFBP3	-0.438982138	p < 1e-4	p < 1e-4
MAP4K4	-0.436176474	p < 1e-4	p < 1e-4
NOTCH1	-0.610467618	0.001	0.013409091
PMS2	0.535342391	0.001	0.013409091
MLH1	0.52374045	0.001	0.013409091
EGFR	-0.515311146	0.001	0.013409091
FOXO3	-0.499303329	0.001	0.013409091
PPP1R15A	-0.487186159	0.001	0.013409091
TADA2B	0.480724824	0.001	0.013409091

The top 20 genes by their median difference in correlation for each gene and all other genes in the p53 pathway gene set between non p53-mutated and p53-mutated breast cancer RNA-seq samples. 1000 permutation samples were generated in order to estimate empirical p-values for each gene, which were then adjusted by the Benjamini-Hochberg method to control the false discovery rate among this set of genes