New AI Model Identifies Genetic Risk Factors for Parkinson’s Disease

A team led by Cleveland Clinic has used advanced AI models to uncover genetic factors in Parkinson’s disease progression and identify existing drugs that could be repurposed for treatment, potentially accelerating the development of new therapies.

In a new study, researchers led by the Cleveland Clinic Genome Center have successfully used advanced artificial intelligence (AI) genetics models to identify genetic factors associated with Parkinson’s disease and pinpoint FDA-approved drugs that could be repurposed for treatment.

The research, published in the journal npj Parkinson’s Disease, employs a systems biology approach. This methodology leverages AI to integrate and analyze diverse forms of data — including genetic, proteomic, pharmaceutical and patient datasets — to uncover patterns that might be missed by conventional analysis.

The study was led by Feixiong Cheng, a CCGC director and an expert in the systems biology field. 

“Parkinson’s disease is the second most common neurodegenerative disorder, right after dementia, but we don’t have a way to stop or slow its progression in the millions of people who live with this condition worldwide; the best we can currently accomplish is managing symptoms as they appear,” first author Lijun Dou, a postdoctoral fellow in Cheng’s Genomic Medicine lab, said in a news release. “There is an urgent need to develop new disease-modifying therapies for Parkinson’s disease.”

Parkinson’s disease progression has been difficult to combat partly because the genetic mutations involved often lie in non-coding regions of DNA, which do not encode proteins but can influence gene function.

“Many of the known genetic mutations associated with Parkinson’s disease are in non-coding regions of our DNA, and not in actual genes,” Dou added. “We know that variants in noncoding regions can in turn impact the function of different genes, but we don’t know which genes are impacted in Parkinson’s disease.”

Using their innovative AI model, the team cross-referenced genetic variants connected to Parkinson’s with multiple brain-specific DNA and gene expression datasets. This approach allowed them to identify which genes are affected by variants in non-coding regions of DNA.

The researchers then integrated these findings with protein and interactome datasets to pinpoint how these genes influence other brain proteins when mutated. They identified several potential risk genes, such as SNCA and LRRK2, which are known to cause brain inflammation when dysregulated.

The study didn’t stop at gene identification. The research team explored whether any existing drugs could target the identified genes, aiming to bypass the typical 15-year span required for developing and approving new medications. 

“Individuals currently living with Parkinson’s disease can’t afford to wait that long for new options as their conditions continue to progress. If we can use drugs that are already FDA-approved and repurpose them for Parkinson’s disease, we can significantly reduce the amount of time until we can give patients more options,” Cheng said in the news release.

By integrating their genetic discoveries with pharmaceutical databases, the team identified multiple candidate drugs. They cross-checked electronic health records and found that patients taking some of the identified drugs, like the cholesterol-lowering drug simvastatin, were less likely to be diagnosed with Parkinson’s disease.

The next step in their research involves laboratory testing of simvastatin, alongside several immunosuppressive and anti-anxiety medications that showed promise.

“Using traditional methods, completing any of the steps we took to identify genes, proteins and drugs would be very resource- and time-intensive tasks,” added Dou. “Our integrative network-based analyses allowed us to speed this process up significantly and identify multiple candidates which ups our chance of finding new solutions.”