MoXGATE: Modality-Aware Cross-Attention for Multi-Omic Gastrointestinal Cancer Subtype Classification

Sajib Acharjee Dip

Track: long paper (up to 6 pages)

Keywords: multimodal, attention, cancer subtype classification, multi-omics integration, cross-attention, modality-aware fusion, deep learning, MoXGATE, genomic data, epigenomic data, transcriptomic data, feature fusion, inter-modality dependencies, learnable modality weights, TCGA datasets, Gastrointestinal Adenocarcinoma (GIAC), Breast Cancer (BRCA), classification accuracy, ablation studies, focal loss, data imbalance, model generalization, interpretability, multi-cancer prediction, biological heterogeneity, personalized treatment, prognostic assessment, multi-head attention, transformer models, neural networks, representation learning, supervised learning, end-to-end training, embedding space, attention mechanism, feature importance, regularization, model robustness, imbalanced classification, transfer learning, domain adaptation, hyperparameter tuning

TL;DR: We propose MoXGATE, a modality-aware deep learning model using cross-attention and learnable weights for multi-omics fusion, achieving 95% accuracy in cancer subtype classification and showing strong generalization across cancer types.

Abstract: Cancer subtype classification is crucial for personalized treatment and prognostic assessment. However, effectively integrating multi-omic data remains challenging due to the heterogeneous nature of genomic, epigenomic, and transcriptomic features. In this work, we propose Modality-Aware Cross-Attention MoXGATE, a novel deep-learning framework that leverages cross-attention and learnable modality weights to enhance feature fusion across multiple omics sources. Our approach effectively captures inter-modality dependencies, ensuring robust and interpretable integration. Through experiments on Gastrointestinal Adenocarcinoma (GIAC) and Breast Cancer (BRCA) datasets from TCGA, we demonstrate that MoXGATE outperforms existing methods, achieving 95\% classification accuracy. Ablation studies validate the effectiveness of cross-attention over simple concatenation and highlight the importance of different omics modalities. Moreover, our model generalizes well to unseen cancer types e.g., breast cancer, underscoring its adaptability. Key contributions include (1) a cross-attention-based multi-omic integration framework, (2) modality-weighted fusion for enhanced interpretability, (3) application of focal loss to mitigate data imbalance, and (4) validation across multiple cancer subtypes. Our results indicate that MoXGATE is a promising approach for multi-omic cancer subtype classification, offering improved performance and biological generalizability.

Submission Number: 46