Efficient Mining of Variants From Trios for Ventricular Septal Defect Association Study

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Ventricular septal defect (VSD) is a severe congenital heart disease with significant infant mortality.
• Early diagnosis via genetic variant identification is crucial for VSD management.
• Current genetic variant mining methods (panel-based, trio-based) have limitations including cost, missing rare variants, and computational intensity.

Purpose of the Study:

• To develop and evaluate a novel, efficient algorithm for mining genetic variants associated with VSD from trio-based sequencing data.
• To address limitations of existing variant detection methods, focusing on speed, memory usage, and identification of novel mutations.

Main Methods:

• Developed a trio-based variant mining algorithm incorporating coupled Bloom Filters for k-mer filtering.
• Implemented statistical error correction and k-mer extension to identify candidate variants.
• Analyzed identified variants against existing databases to pinpoint VSD-associated mutations.

Main Results:

• The novel algorithm demonstrated a 10-fold reduction in candidate coding genes and a 5-fold reduction in lncRNAs compared to single-sequence approaches.
• Achieved a 10x speed increase and 2 orders of magnitude memory reduction over state-of-the-art methods.
• Identified potential VSD-related genetic factors in a VSD trio: CD80, MYBPC3/TRDN combination, and NONHSAT096266.2 lncRNA.

Conclusions:

• The developed algorithm offers a computationally efficient and effective method for identifying VSD-associated genetic variants.
• This approach successfully identified novel and known genetic candidates, including unreported genes and lncRNAs, relevant to VSD.
• The findings provide a foundation for improved genetic diagnosis and understanding of VSD etiology.