Lukas Gerasimavicius

The paper identifies ER-liquid condensates that sequester FAM134B/C and RhoA, affecting cell morphology and cancer invasion.

This paper introduces SPDV, a spatial distance metric that outperforms half of existing variant effect predictors in identifying dominant disease variants, particularly for gain-of-function missense variants.

This paper evaluates non-coding variants' impact on TF binding using AlphaFold 3 and FoldX, showing structural modeling can provide interpretable insights despite lower accuracy than sequence-based methods.

This study evaluates 70 variant effect predictors for RyR1 missense variants and introduces a structural metric, SPDV, to assign ACMG/AMP evidence levels for clinical interpretation.

The paper uses deep mutational scanning to quantify PAX6:DNA binding and predict clinical outcomes of variants, finding most pathogenic mutations affect DNA binding.

The paper compares protein stability predictors with DMS-derived functional scores, finding FoldX and Rosetta show strong correlations, especially when considering intermolecular interactions.

The paper examines how different mutation types affect protein structure and finds that non-LOF mutations are underpredicted by current tools.

This study evaluates 12 protein stability predictors for identifying pathogenic missense mutations and finds FoldX performs best, though structural predictors are less effective than variant effect predictors overall.