Studying the evolution of transcription factor binding events using multi-species ChIP-Seq data.

Area of Science:

• Evolutionary genomics
• Computational biology
• Bioinformatics

Background:

• ChIP-Seq data enables whole-genome transcription factor binding (TFB) profiling across species.
• Understanding TFB evolution is crucial, but lacks robust statistical models.
• Existing models often overlook INDEL disruptions and data ascertainment bias.

Purpose of the Study:

• To develop a novel phylogenetic method for inferring TFB evolution.
• To model the on/off rates of TFB events, focusing on INDEL disruptions.
• To correct for ascertainment bias inherent in ChIP-Seq data.

Main Methods:

• Developed a phylogenetic tree-based model for TFB event rates.
• Masked nucleotide substitutions, focusing on INDEL disruption events.
• Corrected for ascertainment bias by maximizing conditional likelihood.

Main Results:

• The method performs well in model selection and parameter estimation.
• Applied to vertebrate ChIP-Seq data, higher instantaneous transition rates to INDELs were observed in TFB regions compared to non-binding regions.
• Conserved regions showed significantly lower transition rates, as expected.

Conclusions:

• The developed phylogenetic method accurately models TFB evolution, particularly INDEL disruptions.
• Transcription factor binding regions exhibit higher rates of INDEL disruption than non-binding regions.
• The R package TFBphylo is available for implementing this model.