CATVariant: a web server for integrated protein variant interpretation across sequence, structure, population, and clinical evidence.

Efficient interpretation of the structural, functional, and clinical impact of protein variants remains a longstanding challenge. This difficulty arises because evidence needed to interpret variant effects is distributed across sequence, structure, population, clinical, experimental, and literature resources. CATVariant (https://catvariant.khoa.ngo) is an open-access web server that addresses this gap by integrating heterogeneous evidence sources within a unified framework for protein variant interpretation. Given a protein name, protein structure, or list of variants, CATVariant retrieves, harmonizes, and analyzes residue-level information across multiple literature and data resources. The server generates an integrated evidence map linking sequence and structural views to variant tables, structural analyses, experimental measurements, population and disease context, literature associations, and evidence summaries. CATVariant supports rapid transition from a protein-wide overview to evidence-linked evaluation of specific variants within a unified evidence framework. Analyses of 6388 ClinVar-derived review-set variants across 22 proteins illustrate how integrated evidence supports variant prioritization. Across random pathogenic-benign variant pairs, pathogenic variants rank higher in 76% of comparisons. Analysis of variants in the hERG protein further shows how multi-scale evidence highlights vulnerable regions and generates testable hypotheses of channel dysfunction. Together, these analyses demonstrate rapid, transparent, and scalable prioritization of protein variants for mechanistic and experimental follow-up.