We are funded by the National Institutes of Health (NIH) to develop new and novel computational methods for the analysis of big biomedical data. Below is a sample of our active research grants. An overview of our work supported by these grants and others can be found on our research page.
Artificial Intelligence Strategies for Alzheimer’s Disease Research (R01 AG066833, PI – Moore)
The goal of this research program is to develop automated machine learning (AutoML) methods for the multimodal data analysis of Alzheimer’s disease. Data types include genetics, genomics, and imaging. A central goal is to develop an integrated knowledgebase to inform the AutoML methods. The ultimate goal is to identify new drug targets for treating Alzheimer’s disease. We are working closely with the NIA-funded Alzheimer’s Disease Research Centers (ADRCs) to share ideas, discuss solutions, and to collaborate. This project is a collaboration with Drs. Marylyn Ritchie and Li Shen from the University of Pennsylvania who serve as multiple PIs on the grant.
Here is an AutoML paper we published in Bioinformatics from 2020 which provides some background for the project. Here is the paper in JMIR from 2024 reporting the Alzheimer’s knowledgebase we built to inform the AutoML. Here is the paper in Bioinformatics from 2024 reporting our novel large language model method for natureal language queries of the Alzheimer’s knowledgebase.
Knowledge-guided automated machine learning methods for modeling the interaction of HIV with addictive drugs (R01 LM014572, PI – Moore)
This project builds on our user-friendly and open-source Tree-Based Pipeline Optimization (TPOT) platform that represents one of the very first and most widely used open-source AutoML methods. A major benefit of this approach is that it makes machine learning accessible to novice users because it takes the guesswork and complexity out of picking, running, tuning, and optimizing machine learning algorithms and the various pre- and post-processing methods. Bringing this technology to the clinical and translational research communities will open the door to broad adoption of data mining methods for embracing the complexity of the relationship between multimodal substance abuse biomarkers and clinical outcomes such as HIV progression and severity. We propose here novel algorithms to adapt and extend TPOT for the large volumes of clinical data that are being collected on patients infected with HIV at Cedars-Sinai Medical Center in Los Angeles.
Bioinformatics Strategies for Genome-Wide Association Studies (R01 LM010098, PI – Moore)
The goal of this research program is to develop methods for incorporating expert knowledge about functional genomics annotations from sources such as ENCODE into the genetic analysis of complex diseases using computational methods such as AutoML. Do annotations help identify genetic risk factors? What is the best way to incorporate knowledge? How can knowledge inform interpretation of computational models? The project is a collaboration with Dr. Folkert Asselbergs from the University of Utrecht and Dr. Scott Williams from Case-Western Reserve University who both serve as multiple PIs on the grant.
Here is a recent paper published in IEEE/ACM Transactions on Bioinformatics and Computational Biology illustrating our computational approach.
Informatics Algorithms for the Genomic Analysis of Brain Imaging Data (Ro1 LM013463, MPI – Moore)
The goal of this project is develop computational methods, including machine learning, to improve the genetic analysis of brain imaging phenotypes for understanding Alzheimer’s disease. This project addresses the unprecedented scale and complexity of the imaging genetic data sets and the lack of intermediate-level omics data (e.g. transcriptomics) to capture the molecular effects linking genetics to brain phenotypes. This projects is a collaboration with Dr. Li Shen from the University of Pennsylvania who serves as the contact PI for the grant.
Here is a paper we published in Bioinformatics in 2020 which illustrates the type of work this grant funds.