A Machine Learning Approach for Predicting Transcription Factor Binding

• Activity Number: 124
• Type: Topic Contributed
• Date/Time: Monday, August 10, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Steve Qin*
• Companies: Emory University
• Keywords: transcription factor binding ; DNA methylation ; bioinformatics

Abstract

Detecting in vivo transcription factor (TF) binding is important for understanding gene regulatory circuitries. Because inferring genome-wide binding profiles experimentally for multiple TFs is labor-intensive and costly, algorithms for in silico prediction of TF binding have been developed, based mostly on histone modification or DNase I hypersensitivity data in conjunction with DNA motif and other genomic features. However, technical limitations of these methods prevent them from being applied broadly. Exploiting the connection between DNA methylation and TF binding, we proposed a novel supervised learning approach to predict TF-DNA interaction using data from base-resolution whole-genome methylation sequencing experiments (BS-seq and TAB-seq). Comprehensive tests show that the proposed method accurately predicts TF binding and performs favorably versus competing methods.