Prediction of cis-regulatory modules in genomes

Annotating all cis-regulatory modules (CRMs) and constituent transcription factor (TF) binding sites (TFBSs) in genomes is essential to understand genome functions, however, the task remains highly challenging. Here, we developed a new algorithm dePCRM2 for predicting CRMs and TFBSs by integrating numerous TF ChIP-seq datasets. We predicted 1,404,973 CRMCs. And dePCRM2 largely outperforms existing methods. Epigenomic marks play complex roles in cell fate determination. However, little is known about the sequence determinants defining them, we showed two types of convolutional neural networks (CNNs) for cell types and for histone marks are good strategies to uncover the sequence determinants and their importance and interactions. After developed pipeline for predicting the map of CMRs, and a strategy to pinpoint the importance of the motifs in the epigenetic marks in the CRMs, then a complete categorization of cis-regulatory modules (CRMs) and constituent TFBSs in the human and model organismes can facilitate characterizing functions of regulatory sequences in the organisms.To aid the use of these predicted CRMs and TFBSs by the research community, we developed an online database PCRMS (predicted CRMs).The PCRMS database can be a useful resource for the research community to characterize functions of regulatory genomes in important organisms.