The field of biomedical research is witnessing a significant shift towards the integration of multimodal data and the application of self-supervised learning techniques. Researchers are exploring new ways to leverage the inherent structure of unlabeled data to improve predictive accuracy and reduce the dependency on annotated data. This trend is evident in the development of novel frameworks that combine multiple data modalities, such as gene expression data and histopathology images, to enhance the prediction of phenotypes and disease mechanisms. Furthermore, the use of self-supervised learning methods is gaining traction, as they have been shown to outperform traditional supervised models in certain tasks. Noteworthy papers in this area include those that proposed innovative approaches to gene expression prediction, such as the use of dual-pathway multi-level discrimination and bipartite patient-modality graph learning. Additionally, studies that demonstrated the effectiveness of semi-supervised learning in ECG delineation and cancer survival prediction are also noteworthy. These advancements have the potential to revolutionize the field of biomedical research and enable the development of more accurate and personalized treatment strategies.