Uncover transcription factor mediated gene regulations using Bayesian nonnegative factor models

Transcriptional regulation by transcription factors (TFs) controls when and how much RNA is created. Due to technical limitations, the protein level expressions of TFs are usually unknown, making computational reconstruction of transcriptional network a difficult task. We proposed here a novel Bayesian non-negative factor analysis approach, which is capable to estimate both the non-negative abundances of the transcription factors, their regulatory effects, and sample clustering information by integrating microarray data and existing knowledge regarding TFs regulated target genes; further more, we show that the approach can be slightly altered to include miRNA regulations as well. The results demonstrated its validity and effectiveness to reconstructing transcriptional networks by transcription factors through artificial and real data.