Predictive modeling can advance research and facilitate quality improvement initiatives and substantiate research results, especially when data from multiple healthcare systems can be included. However, current, state-of-the-art privacy-preserving predictive modeling frameworks are still centralized, in other words, the models from distributed sites are integrated in a central server to build a global model. This centralization carries several risks, e.g., single-point-of-failure at the central server. To improve the security and robustness of predictive modeling frameworks, we will develop and implement novel and advanced algorithms on decentralized blockchain networks (a distributed ledger/database technology adopted by crypto-currencies such as Bitcoin and Ethereum) to build better models. The outcome will be algorithms that improve the predictive power of data from multiple healthcare systems through a distributed system. Selected references: PMID 36402113, 34923447, and 31943009.

Our goal is to identify potential drug targets of brain disorders (e.g., Alzheimer’s disease) through gene networks comprising disease-associated genes. Recent genomic studies have advanced our knowledge of the genetics of brain disorders and related traits, which could illuminate the pathogenesis of brain disorders. 
The new knowledge provides opportunities for genetic-based strategies for drug target identification. Bioinformatics analyses will be performed to prioritize drug targets and potential drugs for repurposing.

A major challenge hindering progress in neuropsychiatric medicine is our limited understanding of the genetics underlying the complexity of human brain structure and function. Our project aims to characterize genetic effects on the brain by multimodal imaging using human biobanks with MRI and genotype data. This will provide insight into shared and distinct genetic influences among different brain regions. Building on improved genetic knowledge of the brain, we will determine genetic relationship between brain morphology and neuropsychiatric disorders using statistical genetics tools. We will estimate effects of neuropsychiatric genetic risks and environmental exposures on deviations of MRI phenotypes from normal neurodevelopmental and aging trajectories.

Some individuals seem to have lingering or failed recoveries after COVID-19 infections. Students comfortable with basic programming or data science skills are encouraged to enhance our description of recovery profiles from TemPredict, and search for features that can contribute to pre-recovery classification.

To identify molecular mechanisms that contribute to tumor development and maintenance, we develop hypotheses driven computational tools for the integrative analysis of different layers of genetic and epigenetic information. As we recognized that our epigenetic mapping studies can identify effective drug targets, we are now profiling 3D tumor genomes to uncover molecular mechanisms that may cause disturbed enhancer-gene interactions leading to deregulation of gene expression and biochemical pathways.

It remains unclear to what extent chromatin-associated RNAs can reflect the 3D organization of the genome. To this end, we used iMARGI to map genome-wide RNA-chromatin interactions in H1, HFFc6, and K562 cells and yielded on average 1.9 billion read pairs per sample. This project will compare these iMARGI data with genome interaction data including Hi-C and PLAC-seq on three different scales. At the compartment scale, we will test whether the A compartment chromatin is associated with large amounts of RNAs, involving both intrachromosomal and interchromosomal RNA-chromatin interactions. At the TAD scale, we will test whether the RNA ends of nearly all RNA-chromatin interactions are confined to within the boundaries of one or of a few consecutive TADs. At the loop scale, we will test whether RNA-chromatin interactions are enriched with PLAC-seq derived enhancer-promoter interactions.