Genome-wide Prediction of DNase I Hypersensitivity Using Gene Expression

ABSTRACT We evaluate the feasibility of using a biological sample’s transcriptome to predict its genome-wide regulatory element activities measured by DNase I hypersensitivity (DH). We develop BIRD, Big Data Regression for predicting DH, to handle this high-dimensional problem. Applying BIRD to the Encyclopedia of DNA Element (ENCODE) data, we found that gene expression to a large extent predicts DH, and information useful for prediction is contained in the whole transcriptome rather than limited to a regulatory element’s neighboring genes. We show that the predicted DH predicts transcription factor binding sites (TFBSs), prediction models trained using ENCODE data can be applied to gene expression samples in Gene Expression Omnibus (GEO) to predict regulome, and one can use predictions as pseudo-replicates to improve the analysis of high-throughput regulome profiling data. Besides improving our understanding of the regulome-transcriptome relationship, this study suggests that transcriptome-based prediction can provide a useful new approach for regulome mapping.