CMGL: Confidence-guided Multi-omics Graph Learning for Cancer Subtype Classification
For cancer subtype classification, CMGL provides a principled way to handle variable omics quality, improving accuracy and enabling cross-cancer transfer without fine-tuning.
CMGL improves cancer subtype classification by using evidential deep learning to estimate per-sample modality reliability, then using those confidence scores to guide multi-omics fusion and graph construction. It surpasses the strongest baseline by 4.03% average accuracy on four single-cancer tasks and shows transferability to unseen cancer types.
Motivation: Multi-omics integration can improve cancer subtyping, but modality informativeness and noise vary across cancer types and patients. Existing graph-based methods optimize modality weights jointly with the classification objective and therefore lack independent reliability estimates, so low-quality omics distort patient similarity graphs and amplify noise through message passing. Results: We propose CMGL, a two-stage framework that estimates per-sample modality reliability through evidential deep learning and uses the frozen confidence scores to guide cross-omics fusion and graph construction. On four MLOmics cancer-subtype tasks and the 32-class pan-cancer task, CMGL consistently improves over the strongest baseline, surpassing it by 4.03% in average accuracy on the four single-cancer tasks. Its representations recover the PAM50 intrinsic subtypes of breast invasive carcinoma (BRCA), and the BRCA-trained model transfers without fine-tuning to kidney renal clear cell carcinoma (KIRC), stratifying patients into prognostically distinct groups.