BART: a transcription factor prediction tool with query gene sets or epigenomic profiles

Area of Science:

• Genomics
• Computational Biology
• Gene Regulation

Background:

• Identifying functional transcription factors is crucial for understanding gene regulation.
• Traditional methods like DNA motif analysis have limitations in predicting specific factor binding and detecting distal enhancer interactions.

Purpose of the Study:

• To introduce Binding Analysis for Regulation of Transcription (BART), a novel computational method.
• To predict functional transcription factors that regulate specific gene sets or associate with genomic profiles.

Main Methods:

• Developed a computational method and software package named BART.
• Utilized over 6000 ChIP-seq datasets for more than 400 transcription factors in human and mouse.
• Implemented BART in Python for accessibility.

Main Results:

• BART effectively predicts functional transcription factors regulating query gene sets.
• The method demonstrates improved sensitivity for detecting factors binding at distal enhancers compared to conventional approaches.
• Highlights the utility of publicly available ChIP-seq data for functional genomics.

Conclusions:

• BART offers a powerful computational approach for transcription factor prediction in gene regulation.
• The method enhances the analysis of functional genomics by integrating existing large-scale datasets.
• BART provides a valuable tool for researchers studying gene expression and regulation.