Gene-specific artificial intelligence-based variant classification engine: results of a time-capsule experiment

Abstract

Background: Interpretation of genetic variation remains an impediment to cost-effective application of genomics to medicine. An advanced artificial intelligence (AI)-based Variant Classification Engine (aiVCE), rooted in ACMG/AMP guidelines, employs data-driven methods to expedite gene-specific classification (franklin.genoox.com). In this blinded study, the aiVCE’s overall and rule-level performances were evaluated using ClinVar (v. 2018-10) variants with creation dates after 5/01/2017. By removing any prior knowledge of these variants from the aiVCE training data, they were treated as novel variants. Using a ‘Full’ dataset (75,801 variants with ≥1 star) and an ‘Increased-Certainty’ dataset (3,993 variants with ≥2 stars), the aiVCE classified variants as pathogenic (P), likely-pathogenic (LP), uncertain significance (VUS), likely-benign (LB), or benign (B). VUS with sufficient supporting data were subclassified as VUS-leaning benign or VUS-leaning pathogenic. aiVCE results were evaluated to determine concordance with final ClinVar classification and rule-level determinations. Results: The aiVCE demonstrated >97% concordance among Increased-Certainty variants. Concordance was >95% across variant effects (e.g., missense, null, splice region), and was >93.5% for the Full dataset. When assessing the aiVCE’s application of specific ACMG rules, significant differences were observed between ClinVar P/LP and B/LB variants rule-met proportions (all P<0.00001), thus supporting gene-specific rule selections. Evaluation of discordance between the aiVCE and ClinVar uncovered evidences that might have been unavailable to submitting laboratories, highlighting AI utility in variant classification. Conclusions: The aiVCE exhibited robust performance, despite lacking past evidence, in determining whether variants would be categorized as P/LP. Applying latest computational advances to existing guidelines may assist scientists and clinicians interpret variants with limited clinical information and greatly reduce analytical bottlenecks.

A Comprehensive Analysis of Hungarian MODY Patients—Part II: Glucokinase MODY Is the Most Prevalent Subtype Responsible for about 70% of Confirmed Cases

“…The remaining variants were classified according to the ACMG standards and guidelines [ 21 , 22 ]. A web-based interpretation tool, Franklin (Genoox) [ 23 ] was used to assist the classification. HGMD Professional and ClinVar databases were also used in variant interpretation.…”

SCAR: The high‐prevalence antigen 013.008 in the Scianna blood group system

“…Computational prediction by Franklin 36 indicated the ERMAP and RHCE substitutions to be either benign or of uncertain significance (Tables S1 and S2). Computational modeling by PredictSNP 38 also indicated the non‐synonymous ERMAP:p.(Gln142Glu) as neutral with an expected accuracy of 83%.…”

“…Franklin by Genoox (https://franklin.genoox.com/home), 36 an artificial intelligence‐based variant classification and interpretation based on guidelines by the American College of Medical Genetics and Genomics (ACMG), 37 was used to predict the functional impact of sequence variants. PredictSNP was also applied to predict the functional impact of non‐synonymous nucleotide substitutions 38 …”

Four New Cases of Hypomyelinating Leukodystrophy Associated with the UFM1 c.-155_-153delTCA Founder Mutation in Pediatric Patients of Roma Descent in Hungary

“…In the case of WES and WGS data, raw data were aligned to the hg19 reference genome using NextGene software (SoftGenetics, State College, PA, USA), the variant list was uploaded to the Franklin Analysis Platform (Genoox) [ 13 ] for variant classification, and HPO terms were used to help with variant prioritization.…”