Finding potential cis-regulatory loci using allele-specific chromatin accessibility as weights in a kernel-based...

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Identifying cis-acting regulatory elements is crucial for understanding gene expression.
• Existing methods may not fully capture the functional impact of non-coding variants.
• Allele-specific regulation plays a significant role in gene expression variability.

Purpose of the Study:

• To develop and validate a novel computational approach for detecting cis-acting regulatory loci.
• To integrate functional potential estimates (DNase-seq) with variance component association analysis.
• To identify specific single-nucleotide polymorphisms (SNPs) involved in allele-specific cis-regulation.

Main Methods:

• Utilized ENCODE DNase-seq data to estimate functional potential of heterozygous non-coding SNPs.
• Employed kernel-based variance component association analyses on Genetic Analysis Workshop 19 (GAW19) data.
• Constructed weighted and non-weighted covariance kernels to test for cis-regulatory effects against transcript expression phenotypes.

Main Results:

• Identified ten transcripts with potential cis-regulatory effects exceeding the Bonferroni significance threshold.
• Demonstrated that removing cis-located SNPs from the weighted kernel abolished the association signal for the top transcript.
• Provided compelling evidence for allele-specific cis-regulation in four transcripts using both kernel methods.

Conclusions:

• The developed method effectively detects potential cis-acting regulatory loci by integrating functional potential and association analysis.
• Results support the involvement of cis-located variants in the regulation of neighboring genes, consistent with prior research.
• The approach offers a powerful tool for dissecting allele-specific gene regulation mechanisms.