Accurate inference of transcription factor binding from DNA sequence and chromatin accessibility data

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Accurate functional annotation of regulatory elements is crucial for understanding global gene regulation.
• Transcription factors (TFs) play a key role in controlling gene expression by binding to specific DNA sequences.

Purpose of the Study:

• To generate a comprehensive genome-wide map of transcription factor binding sites (TFBS) in human lymphoblastoid cell lines.
• To identify known and novel sequence motifs recognized by transcription factors.

Main Methods:

• Developed a probabilistic framework integrating cell-specific experimental data (histone modifications, DNase I cleavage patterns) with genomic information (gene annotation, evolutionary conservation).
• Applied the framework to create a map of 827,000 TFBS, including sites for 239 known TF binding motifs and 49 novel motifs.

Main Results:

• Generated a high-resolution genome-wide map of 827,000 transcription factor binding sites.
• Identified 239 known and 49 novel sequence motifs associated with TF binding.
• Validated the accuracy of the method by comparison with empirical ChIP-seq data.

Conclusions:

• The developed probabilistic framework provides an accurate and efficient method for genome-wide TFBS mapping.
• This approach enables the simultaneous targeting of multiple factors in a single assay, offering an advantage over traditional methods.
• The generated map is a valuable resource for future genome-wide studies of gene regulation across diverse cell types and conditions.