The field of disease diagnosis and biomarker identification is rapidly advancing with the integration of multi-omics data and graph-based deep learning techniques. Recent studies have demonstrated the potential of these approaches to improve the accuracy and interpretability of disease classification and biomarker identification. The use of graph neural networks and multimodal machine learning models has shown promise in integrating diverse types of data, such as genomic, transcriptomic, and proteomic data, to identify complex patterns and relationships that are associated with disease. Notably, the development of novel models such as MOGKAN and MSNGO has achieved state-of-the-art performance in multi-cancer classification and protein function prediction tasks. Furthermore, the application of these techniques to specific diseases, such as non-small cell lung cancer and Alzheimer's disease, has led to the identification of potential biomarkers and therapeutic targets. Overall, these advances have the potential to transform our understanding of disease mechanisms and improve the diagnosis and treatment of complex diseases. Noteworthy papers include: MOGKAN, which introduces a deep learning model for multi-omics integration and achieves high accuracy in cancer classification; MSNGO, which proposes a novel approach for multi-species protein function prediction using structural features and network propagation; and SCMPPI, which develops a supervised contrastive multimodal framework for predicting protein-protein interactions.