RT Journal Article SR Electronic T1 A Bayesian network approach for modeling mixed features in TCGA ovarian cancer data JF bioRxiv FD Cold Spring Harbor Laboratory SP 033332 DO 10.1101/033332 A1 Qingyang Zhang A1 Ji-Ping Wang A1 Northwestern PSOC members YR 2016 UL http://biorxiv.org/content/early/2016/03/26/033332.abstract AB We propose an integrative framework to select important genetic and epigenetic features related to ovarian cancer and to quantify the causal relationships among these features using a logistic Bayesian network model based on The Cancer Genome Atlas data. The constructed Bayesian network has identified four gene clusters of distinct cellular functions, 13 driver genes, as well as some new biological pathways which may shed new light into the molecular mechanisms of ovarian cancer.