Molecular characterization and pathogenicity stratification of YWHAG missense variants in developmental and epileptic encephalopathy.

MetaSVM, MetaLR, and ESM1b outperform traditional tools in predicting YWHAG missense variant pathogenicity, with MetaSVM showing the best overall performance.

• Why it matters: Accurate classification of YWHAG missense variants is crucial for diagnosing developmental and epileptic encephalopathy (DEE), but current methods lack reliability. Understanding the relationship between YWHAG expression patterns and pathogenicity can improve clinical diagnosis.
• What they did: The study analyzed 55 YWHAG variants, including 17 pathogenic/likely pathogenic and 38 benign, using protein structure modeling, expression profiling, single-cell transcriptomics, and evaluation of 34 pathogenicity prediction tools based on accuracy, AUC, MCC, and other metrics.
• The result: Meta-predictors like MetaSVM effectively predict YWHAG variant pathogenicity, and YWHAG’s expression pattern during development supports its role in DEE, aiding clinical genetic diagnosis.