Dr. Chang Su, assistant professor of population health sciences and Walsh McDermott Scholar in Public Health, has received an R01 to study Parkinson’s disease (PD) using integrated artificial intelligence (AI) approaches.
PD is the second most prevalent neurodegenerative disorder worldwide, affecting two to three percent of people over the age of 65. PD patients experience a range of debilitating motor impairments, including rigidity, bradykinesia, tremors, postural instability, and gait disorders. They also usually develop non-motor dysfunctions like cognitive impairment, autonomic dysfunction, psychological disorders, sleep disorders, and sensory dysfunctions, significantly impacting their quality of life. Currently, there is no disease-modifying treatment capable of reversing or halting PD progression.
“The progression of PD varies significantly across different symptom dimensions among patients, posing substantial challenges for clinical management and trial design,” explained Dr. Su. “The availability of diverse multimodal biomedical data, which encodes patient health profiles from various perspectives and scales, offers unprecedented opportunities to drive advancements in the field.”
The research team is developing a deep learning model that will integrate multimodal information to capture the heterogeneous progression of PD patients and gain deeper insights into treatment development. The model will also identify clinically meaningful disease subtypes with distinct progression patterns and trajectories.
They will leverage longitudinal clinical records, multi-omics, and neuroimaging acquisitions from two multi-institutional research cohorts: the Parkinson’s Progression Markers Initiative and the Parkinson’s Disease Biomarkers Program.
The team will next seek to identify the molecular drivers governing those PD subtypes. “By integrating individual multi-omics data with functional genomics and human protein interactome,” said Dr. Su, “we aim to construct subtype-specific gene modules that drive different PD progression trajectories, which can be targeted for novel therapeutic development.”
The researchers will further conduct in-silico drug repurposing, predicting potential drug candidates that were previously approved for other diseases and can be tailored to PD subtypes. They will first examine individual-level transcriptomics and drug-perturbation gene expression profiles in cultured cell lines to generate drug-repurposing hypotheses, targeting the subtype-specific gene modules. Next, they will predict drug repurposing candidates by leveraging inference techniques within a comprehensive biomedical knowledge graph they have built. Finally, to assess these drug candidates, the team will conduct real-world data (RWD)-based trial emulation, drawing real-world evidence from three large-scale repositories: INSIGHT Clinical Research Network (CRN), Cleveland Clinic, and Temple University.
Dr. Su is working with Dr. Fei Wang, director of the Institute of Artificial Intelligence for Digital Health at Weill Cornell Medicine (AIDH) and professor of population health sciences, Dr. Harini Sarva, director of the Parkinson's Disease & Movement Disorders Institute and associate professor of clinical neurology, Dr. Yifan Peng, associate professor of population health sciences, and Dr. Wodan Ling, assistant professor of population health sciences. He will also work with Dr. Feixiong Cheng, director of the Cleveland Clinic Genome Center, Dr. Xinghua Shi, associate professor in the Department of Computer & Information Sciences at Temple University, and Dr. Molly Cincotta, assistant professor of clinical neurology at Temple University. He extends gratitude to members of the Department of Population Health Sciences, the Division of Health Informatics, AIDH, INSIGHT, and the grants and finance team for their continued support.
“Holistically analyzing diverse biomedical data with integrated AI offers a tremendous opportunity to deepen our understanding of PD progression heterogeneity and accelerate the development of effective treatments,” Dr. Su concluded. “Our goal is to establish a scalable computational pipeline that not only advances PD research but can be extended to tackle other complex human diseases.”
- Highlights