Bayesian Hierarchical Model of Protein Binding Microarray K-mer Data

• Activity Number: 466
• Type: Invited
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: International Society for Bayesian Analysis (ISBA)
• Author(s): Jun S. Liu*+
• Companies: Harvard University
• Keywords: bayesian modeling ; transcription factors ; high-throughput ; genome-wide

Abstract

In recent years numerous studies have employed universal protein binding microarray (PBM) technology to determine the in vitro DNA binding specificities of hundreds of TFs for all possible 8-bp sequences (8-mers). We have developed a Bayesian hierarchical analysis of variance (ANOVA) model that decomposes these 8-mer data into background noise, TF family-wise effects, and effects due to the particular TF. Adjusting for the background noise improves PBM data quality and concordance with in vivo TF binding data. Moreover, our model provides simultaneous identification of TF families and their shared DNA binding preferences, and also identification of 8-mers bound preferentially by individual members of a TF family. Such modeling may aid in deciphering cis-regulatory codes in the genome and understanding the molecular determinants of protein-DNA binding specificity.