Systemic diseases, including cardiometabolic disease (CMD), are often progressive and multifactorial, requiring different biomarkers at each stage for early detection. Treating them effectively often involves polypharmacy (drug combinations) or polypharmacology (multi-targeted drugs). On top of that, patient heterogeneity further complicates treatment. Current data-driven approaches tend to overlook the time dynamics of disease progression and fail to capture its full complexity, limiting their clinical relevance. Incorporating those allows for the discovery of time-specific biomarkers and synergistic drug target combinations to treat different factors within the disease. In this project, we will use systems biology tools and deep learning to integrate transcriptomic, metabolomics, and proteomics data to understand the disease progression and identify new biomarkers for precision medicine purposes. For the deep learning part, we aim to develop foundation models for multi-omics data, similar to single-cell foundation models like the recently developed scGPT [1] and tGPT [2], but with a focus on understanding mechanisms of CMD. [1] Cui, H., Wang, C., Maan, H., Pang, K., Luo, F., Duan, N., & Wang, B. (2024). ScGPT: Toward building a foundation model for single-cell multi-omics using generative AI. Nature Methods, 21, 1470–1480. [2] Shen, H., Liu, J., Hu, J., Shen, X., Zhang, C., Wu, D., Feng, M., Yang, M., Li, Y., Yang, Y., Wang, W., Zhang, Q., Yang, J., Chen, K., & Li, X. (2023). Generative pretraining from large-scale transcriptomes for single-cell deciphering. iScience, 26.