Integrating Multi-Omics Data With EHR for Precision Medicine Using Advanced Artificial Intelligence.

With the recent advancement of novel biomedical technologies such as high-throughput sequencing and wearable devices, multi-modal biomedical data ranging from multi-omics molecular data to real-time continuous bio-signals are generated at an unprecedented speed and scale every day. For the first time, these multi-modal biomedical data are able to make precision medicine close to a reality. However, due to data volume and the complexity, making good use of these multi-modal biomedical data requires major effort. Researchers and clinicians are actively developing artificial intelligence (AI) approaches for data-driven knowledge discovery and causal inference using a variety of biomedical data modalities. These AI-based approaches have demonstrated promising results in various biomedical and healthcare applications. In this review paper, we summarize the state-of-the-art AI models for integrating multi-omics data and electronic health records (EHRs) for precision medicine. We discuss the challenges and opportunities in integrating multi-omics data with EHRs and future directions. We hope this review can inspire future research and developing in integrating multi-omics data with EHRs for precision medicine.

Identifying and characterizing disease subpopulations that most benefit from polygenic risk scores.

Polygenic risk scores (PRSs) hold promise in their potential translation into clinical settings to improve disease risk prediction. An important consideration in integrating PRSs into clinical settings is to gain an understanding of how to identify which subpopulations of individuals most benefit from PRSs for risk prediction. In this study, using the UK Biobank dataset, we trained logistic regression models to predict the 10 year incident risk of myocardial infarction, breast cancer, and schizophrenia using either just clinical features or clinical features combined with PRSs. For each disease, we identified the top 10% subgroup with the greatest magnitude of improvement in risk prediction accuracy attributed to PRSs in the multi-modal model. Using up to ~ 3.6 k demographic, lifestyle, diagnostic, lab, and physical measurement features from the UK Biobank dataset of ~ 500 k individuals, we characterized these subgroups based on various clinical, lifestyle, and demographic characteristics. The incident cases in the top 10% subgroup for each disease represent distinct phenotypes that differ from other cases and that are strongly correlated with genetic predisposition. Our findings provide insights into disease subtypes and can encourage future studies aimed at classifying these individuals to enhance the targeting of polygenic risk scoring in practice.