UNCOVER COOPERATIVE GENE REGULATIONS BY MICRORNAS AND TRANSCRIPTION FACTORS IN GLIOBLASTOMA USING A NONNEGATIVE HYBRID FACTOR MODEL
Presented by: Yufei Huang, Author(s): Jia Meng, The University of Texas at San Antonio, United States; Hung-I Chen, The University of Texas Health Science Center at San Antonio, United States; Jianqiu Zhang, The University of Texas at San Antonio, United States; Yidong Chen, The University of Texas Health Science Center at San Antonio, United States; Yufei Huang, The University of Texas at San Antonio, United States
The problem of uncovering regulatory networks (TRNs) by transcription factors (TFs) and microRNAs (miRNAs) is considered. A novel Bayesian hybrid factor analysis (BHFA) approach is proposed that includes a hybrid factor and regression model to capture the cooperative regulations of miRNA and TFs and a Gibbs sampling solution for infer a context-specific transcriptional regulatory network as well as the TF protein level expression profiles. Particularly, the proposed model capture the correlated and context-specific regulations explicitly using a hierarchical Dirichlet process (HDP) mixture model. It also considers the sparsity of transcriptional regulation, the non-negative TF activities, and the down-regulatory effects of miRNAs, and admits the prior knowledge regarding TF and miRNA regulated target genes. The proposed BHFA approach is tested on both simulated data and glioblastoma multiforme (GBM) microarray data. The results demonstrate the validity and effectiveness the proposed BHFA approach for uncovering the TRNs.