Creating and validating cis-regulatory maps of tissue-specific gene expression regulation

Area of Science:

• Genomics
• Computational Biology
• Gene Regulation

Background:

• Identifying genomic regions that control gene transcription is a significant challenge in molecular biology.
• Existing methods for mapping cis-regulatory modules (CRMs) to genes have limitations in accuracy and scope.

Purpose of the Study:

• To develop and validate a novel computational approach for creating accurate maps of cis-regulatory modules (CRMs) and their associated genes.
• To infer regulatory relationships that explain gene expression patterns across different tissues.

Main Methods:

• Inferred regulatory relationships by correlating histone modifications at CRMs with gene expression in target tissues.
• Predicted CRM regulatory targets using cross-tissue correlations between histone modifications and gene expression within a 1 Mbp window.
• Validated the accuracy of the generated cis-regulatory maps by comparing gene expression models against control maps and nearest-neighbor heuristics.

Main Results:

• The novel cis-regulatory maps generated more accurate gene expression models compared to control maps.
• The method successfully identified long-range regulatory interactions, outperforming existing mapping strategies.
• Including CRMs predicted across multiple tissues improved map-building accuracy.
• H3K27ac histone modification and CAGE gene expression measurements were identified as the most informative for map creation.

Conclusions:

• The developed computational approach provides a robust framework for building and validating cis-regulatory maps.
• The findings highlight the importance of multi-tissue data integration and specific epigenetic marks for understanding gene regulation.
• This method offers a significant advancement in predicting gene transcription control by genomic regions.