HOCOMOCO: expansion and enhancement of the collection of transcription factor binding sites models

Abstract

Models of transcription factor (TF) binding sites provide a basis for a wide spectrum of studies in regulatory genomics, from reconstruction of regulatory networks to functional annotation of transcripts and sequence variants. While TFs may recognize different sequence patterns in different conditions, it is pragmatic to have a single generic model for each particular TF as a baseline for practical applications. Here we present the expanded and enhanced version of HOCOMOCO (http://hocomoco.autosome.ru and http://www.cbrc.kaust.edu.sa/hocomoco10), the collection of models of DNA patterns, recognized by transcription factors. HOCOMOCO now provides position weight matrix (PWM) models for binding sites of 601 human TFs and, in addition, PWMs for 396 mouse TFs. Furthermore, we introduce the largest up to date collection of dinucleotide PWM models for 86 (52) human (mouse) TFs. The update is based on the analysis of massive ChIP-Seq and HT-SELEX datasets, with the validation of the resulting models on in vivo data. To facilitate a practical application, all HOCOMOCO models are linked to gene and protein databases (Entrez Gene, HGNC, UniProt) and accompanied by precomputed score thresholds. Finally, we provide command-line tools for PWM and diPWM threshold estimation and motif finding in nucleotide sequences.

Figure 1.

Overview of the HOCOMOCO update workflow.

Figure 2.

Results of the benchmark assessing performance of different model collections on human and mouse ChIP-Seq data. The full height of each bar depicts the total number of assessed TFs for a particular case. The green fraction of a bar depicts the number of TFs for which a model from the given collection was the best (had the highest wAUC). The white fraction of the mono-HOCOMOCO v10 bar consists of TFs with the best models found in the databases that did not participate in the collection assembly (HOMER, SWISSREGULON, JASPAR and the published version of HT-SELEX).

Figure 3.

Coverage of TF structural families by TFBS models of HOCOMOCO v10. The area of each blue circle is proportional to the total number of members of a particular family; the orange smaller circle depicts the fraction of TFs for which TFBS models are available. The TF classification is given according to TFClass.

Figure 4.

Binding models of human and mouse STAT1 TFs. LOGO representations of selected models learned from different ChIP-Seq datasets are shown. wAUC values of different models within species are extremely close (about 0.89 for human and 0.78 for mouse). HOCOMOCO v9 human model is shown as the reference. One of human ChIP-Seq datasets yielded a mouse-like model subtype.