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

2025-11-12

AI Summary

Variant interpretation in prostate cancer genomics faces critical bottlenecks—including insufficient evidence for classifying variants of uncertain significance (VUS), lack of cancer-type-specific benchmarks, and inconsistent cross-database annotations. Method: We introduce Prostate-VarBench, the first prostate-cancer-specific variant benchmark, integrating COSMIC, ClinVar, and TCGA-PRAD data; correcting VEP functional annotation errors; standardizing 56 clinical features; and enabling patient- and gene-aware data splitting. We further propose a sparse-mask-driven, interpretable TabNet model that incorporates AlphaMissense pathogenicity scores for missense variants to support case-level reasoning. Contribution/Results: On an independent test set, our model achieves 89.9% accuracy with balanced performance across all ACMG/AMP classification categories. VEP correction reduces the VUS rate by an absolute 6.5%, markedly enhancing clinical interpretability and diagnostic reliability.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Interpretability, Explainability, and Transparency
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web

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.

Problem

• Classifying prostate cancer variants with uncertain clinical significance
• Addressing inconsistent annotations across genomic data sources
• Developing interpretable AI models for variant pathogenicity prediction

Innovation

• Integrated prostate-specific benchmark with harmonized genomic datasets
• Standardized interpretable features across eight clinical tiers
• Interpretable TabNet model with step-wise sparse masks