Fig. 2From: Modeling transcriptional regulation using gene regulatory networks based on multi-omics data sourcesGRN based prediction models outperform those built using TEPIC affinity scores. a and b correspond to prediction performance for 20 random sets of 1729 GM12878 TGs while C and D were obtained from 1892 K562 TGs. Prediction performances for models corresponding to different inputs were compared using Wilcoxon rank sum test (*** − p < 0.0001,** − p value < 0.001, * − p value < 0.05, ns-not significant)Back to article page