We will develop patient-specific digital twin models by integrating organ-on-chip, organoid, and ex vivo experimental platforms with drug screening and multi-omics profiling. These complementary datasets will be incorporated into AI-driven computational models to capture individual biological characteristics and predict therapeutic responses. By virtually evaluating candidate treatments and identifying the most promising options, the digital twin framework aims to support personalized drug selection, improve treatment efficacy, and reduce trial-and-error in therapeutic decision-making.
Projects
- Patient stratification using drug sensitivity data from cancer patient samples: We will develop model-based clustering methods for identification of patient subgroups that are differentially responsive to targeted therapies.
- Data integration on drug target interaction profiles: We will leverage the compound-protein bioactivity data for understanding compounds’ therapeutic efficacy and side effects..
- Quantitative pharmacology modeling for drug combinations: We will utilize network pharmacology modeling approaches to integrate pharmacological and molecular biology data to understand the mechanisms of action of drug combinations.
- Statistical methods for evaluating drug combination dose-response data: We will develop robust statistical methods for the initial validation of the most promising combination therapies.