Discovering hotspots in functional genomic data superposed on 3D chromatin configuration reconstructions

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Genome spatial organization impacts cellular functions, including gene regulation.
• Previous studies focused on 3D co-localization of discrete genomic features.
• Continuous genomic data, like ChIP-seq peak heights, lacked similar 3D spatial analysis.

Purpose of the Study:

• To develop and validate a method for identifying 3D hotspots from continuous genomic data.
• To assess the utility of patient rule induction method (PRIM) for analyzing 3D genome reconstructions.
• To discover novel regulatory interactions by pinpointing functional 3D hotspots.

Main Methods:

• Applied bump hunting via PRIM to ChIP-seq data overlaid on 3D genome reconstructions of Saccharomyces cerevisiae.
• Utilized k-Nearest Neighbor (k-NN) regression as an alternative algorithm for hotspot identification.
• Employed resampling techniques to assess the significance of inter-gene distances within 3D reconstructions.

Main Results:

• Successfully identified 'functional 3D hotspots' characterized by significantly elevated mean ChIP-seq peak heights.
• Discovered a top hotspot for the transcription factor Swi6 containing known target genes MSB2 and ERG11 on different chromosomes.
• Validated hotspot stability across PRIM parameter settings and concordance with k-NN regression results.

Conclusions:

• The developed analytic approach effectively identifies functional 3D hotspots in the genome.
• This method can reveal novel, potentially long-range, gene regulatory interactions.
• Findings highlight the importance of 3D genome architecture in gene regulation.