Prostate-VarBench: A Benchmark with Interpretable TabNet Framework for Prostate Cancer Variant Classification

TL;DR

This paper introduces Prostate-VarBench, a comprehensive prostate-specific variant benchmark, and develops an interpretable TabNet model that improves prostate cancer variant classification accuracy and reduces uncertain classifications.

Contribution

It creates a harmonized, prostate-specific variant dataset and demonstrates an interpretable machine learning model for better classification of prostate cancer variants.

Findings

Achieved 89.9% accuracy on test data

Reduced VUS by 6.5% through data correction

Provided interpretable model rationales aligned with clinical review

Abstract

Variants of Uncertain Significance (VUS) limit the clinical utility of prostate cancer genomics by delaying diagnosis and therapy when evidence for pathogenicity or benignity is incomplete. Progress is further limited by inconsistent annotations across sources and the absence of a prostate-specific benchmark for fair comparison. We introduce Prostate-VarBench, a curated pipeline for creating prostate-specific benchmarks that integrates COSMIC (somatic cancer mutations), ClinVar (expert-curated clinical variants), and TCGA-PRAD (prostate tumor genomics from The Cancer Genome Atlas) into a harmonized dataset of 193,278 variants supporting patient- or gene-aware splits to prevent data leakage. To ensure data integrity, we corrected a Variant Effect Predictor (VEP) issue that merged multiple transcript records, introducing ambiguity in clinical significance fields. We then standardized 56 interpretable features across eight clinically relevant tiers, including population frequency, variant type, and clinical context. AlphaMissense pathogenicity scores were incorporated to enhance missense variant classification and reduce VUS uncertainty. Building on this resource, we trained an interpretable TabNet model to classify variant pathogenicity, whose step-wise sparse masks provide per-case rationales consistent with molecular tumor board review practices. On the held-out test set, the model achieved 89.9% accuracy with balanced class metrics, and the VEP correction yields an 6.5% absolute reduction in VUS.