Linkage analysis has traditionally functioned as a method to identify the causative gene in monogenic diseases. However, the prevalent nuclear family structure in contemporary society hinders the acquisition of DNA sequencing data for families affecte...
Linkage analysis has traditionally functioned as a method to identify the causative gene in monogenic diseases. However, the prevalent nuclear family structure in contemporary society hinders the acquisition of DNA sequencing data for families affected by rare diseases. This study proposes a novel approach—a gene-based association study—to overcome the limitations posed by linkage analysis in identifying the causative gene for monogenic diseases.
To demonstrate the method on BRCA1/2 genes, we obtained the exome data from gnomAD and utilized a case group consisting of 1,751 breast and ovarian cancer patients with BRCA1/2 data from 2022 at SMC. Employing Variant Effect Predictor (VEP) from Ensembl, we annotated four bioinformatics tools (CADD, BayesDel, REVEL, and VEST4) as pathogenicity predictors.
We annotated four bioinformatics tools (CADD, BayesDel, REVEL, and VEST4) as pathogenicity predictors and assigned scores for variants excluded from VEP results based on molecular consequences. Utilizing ClinVar review status, we established thresholds for these tools. Odds ratios were calculated to compare effect sizes between tools.
Despite varied results, VEST4 consistently demonstrated a modest effect size with odds ratios of 3.130–4.881, showing statistical significance in BRCA1/2. Besides the results of 2021 BRCA1, other tools have exhibited statistical significance as well. The overall analysis suggests the potential of gene-based association studies.