| Title: | MULTI: modeling human aging and disease via multi-organ and multi-omics biomedical data |
| Lead Institution: | Columbia University |
| Principal investigator: |
Application 647044
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
The MULTI project aims to advance our understanding of human aging and disease by integrating multi-organ and multi-omics biomedical data to construct a holistic framework for precision medicine. Human aging and complex diseases are influenced by interconnected biological processes across organ systems and molecular scales, yet existing research often focuses on isolated datasets or single-organ studies. To address this gap, the MULTI consortium leverages large-scale resources like the UK Biobank, a pivotal dataset offering extensive imaging, genetics, proteomics, and other omics data across diverse organ systems and populations. Our overarching goal is to develop AI-driven novel phenotypes that quantify individual susceptibility to aging and disease, providing a foundation for personalized health interventions, disease risk prediction, and population selection for clinical trials. Specific aims include: i) consolidating and harmonizing multi-omics, multi-organ data from the UK Biobank with other resources; ii) investigating the phenotypic, proteomic, and genetic interactions across key organ systems such as the brain, heart, and eye; iii) constructing AI/ML-powered models for individual-level modeling; and iv) validating these novel phenotypes in predicting systemic disease outcomes, cognitive decline, and mortality. By bridging gaps across organ systems and omics scales, the MULTI project seeks to transform the study of aging and disease with reproducible methods and public knowledge-sharing tools to the community. We have the following dissemination and AI training plans per UK Biobank policy: i) we will return derived AI biomarkers to UK Biobank, ii) we will not use any public AI models (e.g., ChatGPT) to fit raw data from UK Biobank, and all AI training will be performed on UKB-RAP, iii) we will dessiminate our scientific findings at our public portals, but no raw data or traceable AI pre-trained models will be made publicly available.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 16031 | Refining the generation, interpretation and application of multi-organ, multi-omics biological aging clocks | Junhao Wen | 2025 | Nature Aging |
Enabling scientific discoveries that improve human health