HEMERA: A Human-Explainable Transformer Model for Estimating Lung Cancer Risk using GWAS Data

2025-10-08

AI Summary

This study addresses the challenge of genetic risk prediction for lung cancer in never-smokers. We propose an end-to-end interpretable Transformer model that learns risk signals directly from raw GWAS genotype data—without requiring clinical covariates. Methodologically, we design a neural genotype embedding coupled with additive positional encoding to accommodate high-dimensional, sparse SNP data; introduce Layer-wise Integrated Gradients for SNP-level attribution to enhance biological interpretability; and apply a functionally informed SNP filtering strategy to refine input representation. Trained on 27,254 participants, the model achieves an AUC of 99.1%. Critically, top attributed SNPs strongly colocalize with established lung cancer susceptibility loci (e.g., *CHRNA3*, *TERT*), validating biological plausibility. Our approach establishes a novel paradigm for precision screening in never-smokers and facilitates hypothesis generation regarding disease mechanisms.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Lung cancer (LC) is the third most common cancer and the leading cause of cancer deaths in the US. Although smoking is the primary risk factor, the occurrence of LC in never-smokers and familial aggregation studies highlight a genetic component. Genetic biomarkers identified through genome-wide association studies (GWAS) are promising tools for assessing LC risk. We introduce HEMERA (Human-Explainable Transformer Model for Estimating Lung Cancer Risk using GWAS Data), a new framework that applies explainable transformer-based deep learning to GWAS data of single nucleotide polymorphisms (SNPs) for predicting LC risk. Unlike prior approaches, HEMERA directly processes raw genotype data without clinical covariates, introducing additive positional encodings, neural genotype embeddings, and refined variant filtering. A post hoc explainability module based on Layer-wise Integrated Gradients enables attribution of model predictions to specific SNPs, aligning strongly with known LC risk loci. Trained on data from 27,254 Million Veteran Program participants, HEMERA achieved >99% AUC (area under receiver characteristics) score. These findings support transparent, hypothesis-generating models for personalized LC risk assessment and early intervention.

Problem

• Predicting lung cancer risk using genetic data from GWAS
• Applying explainable transformer models to raw genotype data
• Identifying specific genetic variants contributing to cancer risk

Innovation

• Transformer model processes raw genotype data directly
• Uses additive positional encodings and neural embeddings
• Implements post hoc explainability with integrated gradients