Inferring CTCF binding patterns and anchored loops across human tissues and cell types

Abstract

AbstractCCCTC-binding factor (CTCF) is a transcription regulator which is involved in many cellular processes. How CTCF recognizes DNA sequence to exert chromosome barrier or enhancer blocking effects remains to be fully interrogated. Despite many computational tools were developed to predict CTCF-mediated loops qualitatively or quantitatively, few could specially evaluate the regulatory potential of DNA sequence at CTCF binding sites (CBSs) and how it affects chromatin loop formation. Here, we developed a deep learning model, DeepAnchor, to precisely characterize the binding patterns for different types of CBSs. By incorporating base-wise genomic/epigenomic features, we revealed distinct chromatin and sequence features for CTCF-mediated insulation and looping at a high resolution, such as two sequence motifs flanking the core CTCF motif at loop-associated CBSs. Besides, we leveraged the predicted anchor score to optimize the loop extrusion model and achieved the best performance in predicting CTCF-anchored loops. We established a compendium of context-specific CTCF-anchored loops across 52 human tissue/cell types and found that genomic disruption of CTCF-anchored loops may represent a general causal mechanism of disease pathogenesis. These computational models, together with the established resource, could facilitate the mechanistic research on how the CTCF-mediatedcis-regulatory elements (CREs) shapes context-specific gene regulation in cell development and disease progression.