We have unlock the predictive potential of retinal images for genetic discovery and disease risk assessment across more than 750 conditions. Our approach not only outperforms traditional methods in predicting diseases but also uncovers new genetic associations. We believe this fusion of AI and genetics will transform target identification and prioritization.

The surge in biomedical data from wearable sensors, electronic health records, and genomics is revolutionizing healthcare, offering significant health improvement opportunities while posing challenges in data management and analysis. To meet these challenges, we developed a tool utilizes state-of-the-art security and scalability technologies to provide an end-to-end solution for big biomedical data analytics.

Obesity contributes substantially to the global burden of disease. We studied how genetics influence obesity using UK Biobank data and found certain rare gene variations impact obesity differently in men, women, and children showing how these variations affect processes like cell death and DNA damage.

The timing of puberty affects future health outcomes. We investigated the underlying biological mechanisms, performing multi-ancestry genetic analyses in ∼800,000 women, identifying 1,080 independent signals associated with puberty timing.

Current medicine lacks data-driven guidance, underutilizing the predictive value of medical histories for diseases. We explored the potential of the medical history to inform on the phenome-wide risk of onset for 1,883 disease endpoints across clinical specialties.

Risk stratification is critical for the identification of high-risk individuals. Here we show the potential of NMR metabolomic profiles to predict multidisease onset across 24 common conditions, including metabolic, vascular, respiratory, musculoskeletal and neurological diseases and cancers.