| Title: | Harnessing Multi-Omics and Multimodal Data to Decipher the Interplay Between Diet, Inflammation, Metabolic Dysregulation, Genetic risk factors, environmental factors and outcome of stroke |
| Lead Institution: | Southern Medical University |
| Principal investigator: |
Application 242473
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
According to the 2019 Global Burden of Disease report, stroke is the second leading cause of death worldwide, accounting for 11.6% of total deaths between 1990 and 2019, and also characterized by a high disability rate. Interestingly, certain aspects of our metabolities, like Homocysteine which is related to vascular injury, carries clues of some of these conditions. Previous observation studys and secondary analysis had showed the link between diet, inflammatory factors, metabolic dysregulation, genetic risk factors, environmental factors and the outcomes of stroke. However, whether there are causal relationships between stroke and factors mentioned before remains unclear. How to predict who will get stroke in the future is still worth exploring.The existing prediction methods are just based on age, past medical history, medication history and other general health information, and are not always accurate. We thus plan to use genetic data to find out direct cause-and-effect relationship between diet, validation, metabolic dysregulation, genetic risk factors, environmental factors and stroke by Mendelian randomization. And then create a new predictive model by using deep learning. Therefore, our study purposes are: 1. Creating a new predictive model to study stroke by using multi-omics and multimodal data including diet, validation, metabolic dysregulation, genetic risk factors and environmental factors. 2. Intending to understand the causal relationship between all kinds of factors above and stroke. 3. Using deep learning to predict who might get stroke in the future using various of measurements.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 18424 | Explainable machine learning integrating biochemical and metabolomic biomarkers with conventional clinical factors improves chronic kidney disease prediction and risk stratification | Jing Ma (+8) | 2026 | BMC Nephrology |
Enabling scientific discoveries that improve human health