By Rachel Tompa, Ph.D. / Allen Institute

It turns out there is a limited set of pathways diseases can take through the brain. When viewed through the lens of their genes, brain diseases fall into some surprising patterns, a new study has found. The study authors say this way of sorting brain diseases and psychiatric disorders could potentially identify new targets for treatments.  

In the study, published in April in the journal PLOS Biology, scientists sorted 40 different brain diseases and disorders by looking at where genes important to those diseases are switched on, or expressed, in the healthy human brain. The Venn diagram of disease-linked genes and their locales revealed some surprising insights — for example, in this analysis, multiple sclerosis looks a lot like brain cancer. 

“We thought that we might be able to understand these disorders better, and to better understand how different disorders relate to each other, based on where these genes are expressed,” said Yashar Zeighami, Ph.D., an assistant professor at McGill University who led the study along with Allen Institute Investigator Michael Hawrylycz, Ph.D. 

To classify brain diseases by gene expression locale, the scientists turned to one of the Allen Institute’s classic resources, the Allen Human Brain Atlas. First released in 2010, this atlas surveys gene expression across the entire healthy adult human brain, pinpointing where in the brain genes are switched on for nearly all genes in the human genome.  

 “It seemed like a natural question: What would gene expression tell us about disease?” said Hawrylycz, who was also one of the lead scientists involved in the creation of the Allen Human Brain Atlas. “In our world, it was a very natural perspective from which to try to classify these diseases.”  

From genes to drug targets 

The researchers identified 40 brain diseases and psychiatric disorders for which their genetic causes are at least partially known, including neurodegenerative diseases like Alzheimer’s and Parkinson’s disease; psychiatric disorders such as autism, schizophrenia, and bipolar disorder; brain cancers such as glioblastoma; and several other brain-related diseases. Most of these diseases and disorders are very complicated, genetically speaking. Some disorders arise from mutations in hundreds of different genes, and genetic causes for the same disease often vary from person to person.  

 The scientists took the genes associated with each disease and asked where in the healthy human brain those genes are expressed. The team then grouped the diseases based on their gene expression patterns, ending up with five different categories of disease. Many of these groupings sync with the diseases’ known symptoms — brain cancers clustered together, as did psychiatric disorders. But some diseases lined up in interesting ways: multiple sclerosis and migraine disorders share a grouping with cancers; some addiction disorders overlap with psychiatric diseases, while alcoholism paired with Huntington’s disease and Parkinson’s.  

 The team also asked which individual neurons and other brain cells switch on disease-related genes, using a different Allen Institute dataset from one region of the cortex, the outermost shell of the brain. They found that genes linked to some diseases, like brain cancers and neurodegenerative diseases like Alzheimer’s disease, tend to cluster in cells known as inhibitory neurons, which shut off other neurons’ activity. Other diseases, including many psychiatric disorders, clustered more with excitatory neurons, those that activate other neurons. Eventually, they want to look at disease-related genes at the level of individual cells across the entire brain, Hawrylycz said.  

 This study gets the team one step closer to pinning down the circuits involved in brain disease, said Thomas Nickl-Jockschat, M.D., an associate professor of psychiatry at the University of Iowa and one of the co-authors of the study. And one step closer to finding new drug targets for diseases where better therapies are sorely needed. 

“If we understand which molecular pathways are most likely impacted by these diseases, we can look for wholly new drug targets,” Nickl-Jockschat said. “Now we can better understand how these diseases emerge, how genetic risk actually mediates vulnerability in certain brain regions and cells, and we can use that knowledge to ultimately hit the disease where it hurts the disease the most.”