DiscoVari: A Web-Based Precision Medicine Tool for Predicting Variant Pathogenicity in Cardiomyopathy- and Channelopathy-Associated Genes

Abstract

Background: With genetic testing advancements, the burden of incidentally identified cardiac disease-associated gene variants is rising. These variants may carry a risk of sudden cardiac death, highlighting the need for accurate diagnostic interpretation. We sought to identify pathogenic hotspots in sudden cardiac death-associated genes using amino acid-level signal-to-noise (S:N) analysis and develop a web-based precision medicine tool, DiscoVari, to improve variant evaluation.

Methods: The minor allele frequency of putatively pathogenic variants was derived from cohort-based cardiomyopathy and channelopathy studies in the literature. We normalized disease-associated minor allele frequencies to rare variants in an ostensibly healthy population (Genome Aggregation Database) to calculate amino acid-level S:N. Amino acids with S:N above the gene-specific threshold were defined as hotspots. DiscoVari was built using JavaScript ES6 and using open-source JavaScript library ReactJS, web development framework Next.js, and JavaScript runtime NodeJS. We validated the ability of DiscoVari to identify pathogenic variants using variants from ClinVar and individuals clinically evaluated at the Duke University Hospitals with cardiac genetic testing.

Results: We developed DiscoVari as an internet-based tool for S:N-based variant hotspots. Upon validation, a higher proportion of ClinVar likely pathogenic/pathogenic variants localized to DiscoVari hotspots (43.1%) than likely benign/benign variants (17.8%; P<0.0001). Further, 75.3% of ClinVar variants reclassified to likely pathogenic/pathogenic were in hotspots, compared with 41.3% of those reclassified as variants of uncertain significance (P<0.0001) and 23.4% of those reclassified as likely benign/benign (P<0.0001). Of the clinical cohort variants, 73.1% of likely pathogenic/pathogenic were in hotspots, compared with 0.0% of likely benign/benign (P<0.01).

Conclusions: DiscoVari reliably identifies disease-susceptible amino acid residues to evaluate variants by searching amino acid-specific S:N ratios.

Keywords: channelopathy; exome sequencing; long QT syndrome; probability; syncope.

Figure 1.

Summary of study cohorts and methodology. (A) Schematic of derivation of disease-associated variants. Only TTN truncating variants were included in CM (blue). ACM, CPVT, LQTS, and HCM variants (orange) were restricted to missense and single amino acid indels. Variant pathogenicity was assessed in ClinVar and by in silico algorithms. (B) Diagram of methodology for gnomAD variant inclusion. Variants with a MAF greater than the highest pathologic MAF for that gene in the disease cohort were excluded. (C) Schematic of S:N modeling design. Disease-associated MAF was normalized to gnomAD MAF to establish S:N ratios at the amino acid-level. Pathogenic hotspots were areas with S:N above calculated gene-specific thresholds. AA, amino acid; ACM; arrhythmogenic cardiomyopathy; ACMG, American College of Medical Genetics and Genomics; CM, cardiomyopathy; CPVT, catecholaminergic polymorphic ventricular tachycardia; gnomAD; genome aggregation database; HCM, hypertrophic cardiomyopathy; LB/B, likely benign/benign; LP/P, likely pathogenic/pathogenic; LQTS, long QT syndrome; MAF, minor allele frequency; S:N, signal-to-noise; TTNtvs, truncating variants in TTN; VUS, variant of uncertain significance.

Figure 2.

LP/P ClinVar variants are more commonly found in S:N hotspots compared LB/B variants. (A) S:N validation workflow. CM and channelopathy variants that were not found in our disease cohorts were included. S:N was applied to all LB/B and LP/P variants. (B) Bar graph of the percent of variants located within S:N hotspots for all LB/B and all LP/P variants. (C) Plotted S:N values for LB/B and LP/P variants in hotspots. (D) CM and channelopathy variants with a classification change in ClinVar from initial VUS to LP/P or LB/B (or vice versa) were included. Variants in our disease cohorts were excluded. Variants were assessed using S:N analysis. (E) Bar graph depicting the percent of variants located within S:N hotspots of variants that were reclassified from VUS to LB/B, from LB/B or LP/P to VUS, or from VUS to LP/P. (F) Plotted S:N values for reclassified variants. Labels indicate most recent variant classification. **, P < 0.01. ****, P < 0.0001. LB/B, likely benign/benign; LP/P, likely pathogenic/pathogenic; S:N, signal-to-noise; VUS, variant of uncertain significance.

Figure 3.

Clinically re-evaluated LP/P variants are more likely to fall within S:N hotspots compared to LB/B variants and VUSs. (A) Bar graph of the percent of variants in the clinical Duke cohort located in S:N hotspots stratified by variants classified as LB/B, VUS, and LP/P at time of genetic testing. (B) Plotted S:N values for VUSs and LP/P variants in the clinical cohort. (C) Bar graph of the percent of VUSs located in S:N hotspots stratified by those considered disease-causative, of unknown significance, or suspected benign based on clinical presentation, family history, and co-segregation of variants with disease. **, P < 0.01. ***, P < 0.001. ****, P < 0.0001. ACMG, American College of Medical Genetics and Genomics; LB/B, likely benign/benign; LP/P, likely pathogenic/pathogenic; S:N, signal-to-noise; VUS, variant of uncertain significance.

Figure 4.. Web architecture of the web application, DiscoVari.

The frontend is what user-interface. The backend is where information is processed with programming logic. 256-bit Advanced Encryption Standard stores all data, and each encryption key is encoded with a rotating set of master keys. Primarily JavaScript ES6 programming language has been used to build the frontend and backend of the web application. The application can be accessed using any standard web browser (Safari, Google Chrome, Firefox, etc.) at the web URL (https://discovarilab.duke.edu)

Figure 5.. Flowchart for incorporation of S:N analysis into ACMG criteria when evaluating incidental variants.

Individuals with variants incidentally identified in CM- or channelopathy-associated genes should be referred to a multi-disciplinary center specializing in cardiovascular genetic testing. DiscoVari may be used as one component of a comprehensive clinical evaluation by providing a correlate for ACMG PM1 criteria. ACMG, American College of Medical Genetics and Genomics; CM, cardiomyopathy; FHx, family history; LB/B, likely benign/benign; LP/P, likely pathogenic/pathogenic; PHx, personal history; S:N, signal-to-noise; VUS, variant of uncertain significance.