Detecting non-allelic homologous recombination from high-throughput sequencing data

Area of Science:

• Genomics and Bioinformatics
• Human Genetics
• Molecular Biology

Background:

• Non-allelic homologous recombination (NAHR) is a significant driver of genome rearrangements.
• NAHR is associated with various genetic disorders, highlighting its clinical relevance.
• Accurate detection of NAHR in high-throughput sequencing data remains a substantial technical hurdle.

Purpose of the Study:

• To develop and validate a probabilistic model for detecting NAHR.
• To assess the model's efficacy in identifying NAHR events within low-coverage sequencing data.
• To characterize the nature and distribution of identified NAHR-mediated genomic alterations.

Main Methods:

• Development of a novel probabilistic computational model specifically designed for NAHR detection.
• Application of the model to analyze low-coverage sequencing data from 44 individuals.
• Systematic identification and characterization of NAHR-mediated deletions and duplications across 324 potential loci.

Main Results:

• The probabilistic model successfully identified NAHR-mediated genomic alterations in at least one of 109 potential NAHR loci across the study cohort.
• Identified NAHR events exhibited segregation patterns based on ancestry and were more prevalent in closely spaced repetitive regions.
• Detected NAHR events frequently resulted in gene duplications or pseudogene formation, impacting well-known genes like GBA and CYP2E1.

Conclusions:

• The developed probabilistic model offers a robust method for detecting NAHR in challenging low-coverage sequencing datasets.
• NAHR is a recurrent mechanism contributing to genetic variation, including duplications and deletions, with implications for gene dosage and function.
• The findings underscore the importance of considering NAHR in the interpretation of genomic data, particularly for disease-associated genes.