SyMetrics: An Integrated Machine Learning Model for Evaluating the Pathogenicity of Synonymous Variants in the Human Genome

Linnaeus Bundalian; Martina Strnadová; Felix Garten; Susanne Horn; Udo Stenzel; Denny Popp; Johannes R. Lemke; Saskia Biskup; Björn Schulte; Patrick May; Frank Bösebeck; Antje Garten; Doreen Thor; Angela Schulz; Julia Hentschel; Janet F. Kelso; Torsten Schöneberg; Diana Le Duc

doi:10.1093/nargab/lqaf211

Genetic & Genomic MedicineNAR Genomics and BioinformaticsDec 2, 2025

Peer-reviewed journal article

SyMetrics: An Integrated Machine Learning Model for Evaluating the Pathogenicity of Synonymous Variants in the Human Genome

Linnaeus Bundalian iD, Martina Strnadová iD, Felix Garten, Susanne Horn iD, Udo Stenzel, Denny Popp, Johannes R. Lemke iD, Saskia Biskup iD, Björn Schulte iD, Patrick May iD, Frank Bösebeck iD, Antje Garten iD, Doreen Thor iD, Angela Schulz iD, Julia Hentschel iD, Janet F. Kelso iD, Torsten Schöneberg iD, Diana Le Duc iD

Abstract

Synonymous single nucleotide variants (sSNVs), traditionally seen as neutral, are now recognized for their biological impact. To assess their relevance, we developed SyMetrics, a framework that integrates predictors of splicing, RNA stability, evolutionary conservation, codon usage, synonymous variation effects, sequence properties, and allele frequency. We analyzed all possible sSNVs across the human genome, and our machine-learning model achieved 97% accuracy in distinguishing deleterious from benign variants, with a ROC-AUC of 0.89, outperforming individual predictors. Our estimates indicate that about 1.98 {+/-} 0.17% of sSNVs absent from population databases are damaging (roughly 900, 000 sSNVs), with an odds ratio of 3.87 for deleteriousness compared to common sSNVs (p View larger version (16K): org.highwire.dtl.DTLVardef@12f1627org.highwire.dtl.DTLVardef@57968borg.highwire.dtl.DTLVardef@5cb42borg.highwire.dtl.DTLVardef@38733c_HPS_FORMAT_FIGEXP M_FIG C_FIG

PDF DOI