Resolving diverse protein-DNA footprints from exonuclease-based ChIP experiments

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Chromatin immunoprecipitation (ChIP) sequencing assays identify genomic regions bound by transcription factors (TFs).
• ChIP-exo enhances positional resolution of TF-DNA interactions using lambda exonuclease digestion, generating directional footprints.
• ChIP-exo data represents a mixture of direct and indirect TF binding events.

Purpose of the Study:

• To develop a computational framework, ExoDiversity, for deconvoluting diverse transcription factor binding modes from ChIP-exo data.
• To analyze TF-DNA interactions without prior motif or TF information, automatically learning binding modes.
• To investigate the relationship between TF binding modes, DNA sequence variations, and structural configurations.

Main Methods:

• Implemented a model-based framework (ExoDiversity) to learn a joint distribution of footprints and motifs.
• Applied ExoDiversity to various transcription factors across different organisms and cell types.
• Utilized DNA shape features and conservation signals to characterize discovered binding modes.

Main Results:

• ExoDiversity successfully resolved complex ChIP-exo footprints into distinct binding modes.
• Identified co-factor TF motifs and small nucleotide variations associated with specific binding modes.
• Discovered that different binding modes exhibit distinct DNA shape features and conservation patterns, offering structural and functional insights.

Conclusions:

• ExoDiversity provides a robust method for dissecting heterogeneous transcription factor binding events captured by ChIP-exo.
• The framework reveals novel insights into TF-DNA complex dynamics, including the role of sequence variations and DNA structure.
• This approach enhances the understanding of transcription factor regulation at a genomic level.