The field of graph neural networks and multi-omics integration is rapidly advancing, with a focus on developing innovative methods to model complex biological systems and improve disease classification. Recent research has highlighted the importance of incorporating heterophily and heterogeneity into graph neural network models, as well as the need for interpretable and scalable approaches. Notable papers in this area include Adaptive Heterogeneous Graph Neural Networks, which proposes a heterophily-aware convolution to tackle challenges in modeling heterophily HGs, and MOTGNN, which employs a novel framework for binary disease classification using multi-omics data. Other noteworthy papers include Multi-Omics Analysis for Cancer Subtype Inference via Unrolling Graph Smoothness Priors, which leverages contrastive learning to embed multi-omics data into a unified semantic space, and Sparse Probabilistic Graph Circuits, which introduces a new class of tractable generative models that operate directly on sparse graph representation.