Research News

Alzheimer’s disease experts and patients alike face a frustrating problem: even with more data from across the globe than ever, there’s still limited infrastructure for analyzing this vast amount of data while accounting for the complexity of genetic backgrounds. Researchers from the Cleveland Clinic laboratory of Feixiong Cheng, PhD, aim to close this gap with a new computer-aided drug discovery pipeline.  

Published in Nature Neuroscience, the Cheng Lab’s computer-aided pipeline lets researchers integrate genetic data, health records and other large datasets for research, which is difficult to do manually. The system also uses human brain organoid models and gene editing approaches to further validate computational findings to provide human evidence to be tested in future clinical studies. 

“Our approach is about making better use of the data that already exists, so we can spend less time sorting through possibilities and more time testing what matters the most,” says Dr. Cheng, who is the Dr. Keyhan and Dr. Jafar Mobasseri Endowed Chair for Innovative Research. “Properly leveraging human genetic and genomic data from global populations can help us move toward treatments that make a real difference for patients and families.” 

Unlike other methods, which provide unmanageably long lists of genes, this pipeline gave the Cheng Lab an actionable list of 19 potential drugs to follow up on. 

From the genetics of Alzheimer’s disease to real drug targets 

One of the biggest challenges in studying the genetics of Alzheimer’s disease research is that DNA sequencing reads out long lists of genes associated with the condition. Just because genes are inherited together doesn’t mean those genes cause the same trait – red hair and freckles often go hand in hand, but are caused by different genes. In the same way, not every gene linked to Alzheimer’s disease causes it.  

Any time a researcher gets a readout of potential genes, each gene has to be tested in the lab. This process can take years and cost hundreds of thousands of dollars. 

Dr. Cheng, who directs Cleveland Clinic’s Genome Center, led his team in using computing and multi-omics-based analyses to transform genetic data from millions of people into a manageable list of variants worth testing.  

Analyzing genomic data from different populations separately helps the team prioritize genes that show consistent patterns across populations and identify variants that are more common in specific populations. 

From drug targets to potential new drugs for Alzheimer’s disease 

One variant the team linked to lower Alzheimer’s risk was in the gene EPHX2. This gene stood out across multiple datasets and was later linked to brain inflammation in brain organoid models of Alzheimer's disease. 

Preclinical gene editing and pharmacological experiments in preclinical models showed that stopping EXPH2 improved neuron activity, memory and cognitive performance.  

The next step is turning genetic results into treatments. Instead of starting from scratch, Dr. Cheng’s group further asked whether any existing medications that already target other genes on their shortlist. They analyzed health records and insurance claims from several million individuals to see if people taking certain drugs developed Alzheimer’s less often, again comparing across populations. 

The team highlighted several drugs with encouraging signals, including trazodone and others that act on identified targets. They also saw differences across populations. For example, methotrexate, which targets a gene on the shortlist called DHFR, was linked to lower Alzheimer’s risk in individuals with African ancestry. 

“Because our data spans different populations, we can be more confident that our findings will apply broadly and help as many people as possible,” Dr. Cheng says. “We hope that by combining advanced computing with human brain organoid models, we can offer a more efficient path forward in the next generation of Alzheimer’s disease treatments.” 

Dr. Cheng also says that his team's experiments demonstrate that the Genome Center can pave the way for new drug discovery avenues into diseases beyond Alzheimer's disease.  

“We specifically focused on Alzheimer’s disease, but human genetics and brain organoid models developed here can be applied to other complex neurological disorders,” he says. “We hope our methods can provide a framework to advance the entire field of neurotherapeutic development and human health.”  

About this study 

The present research involved collaboration between the Genome Center, part of Cleveland Clinic Research, Cleveland Alzheimer’s Disease Research Center and the Cleveland Clinic Lou Ruvo Center for Brain Health. Study first author, Yuan Hou, PhD, spearheaded this effort. This work was primarily supported by the National Institute on Aging.