By Ava Dzurenda
About half of Americans will meet the criteria for one or more psychiatric disorders at some point in their lives. Beyond the direct symptoms, these conditions carry heavy burdens, including missed work or school, strained family relationships and financial stress.
To improve diagnoses and treatments for these disorders, researchers at the University of Pittsburgh contributed data to The Psychiatric Genomics Cross Disorders Working Group to better understand the genetics underlying psychiatric disorders. Published in December 2025 in Nature, the study was the first of its kind to use multiple large genetic databases and statistical methods to analyze data from approximately one million people across 14 childhood- and adult-onset psychiatric disorders.
The results of this study help explain why many psychiatric disorders tend to occur together, showing that they often share underlying genetic risk. By uncovering common biological pathways across disorders, the results may also guide future efforts to develop treatments that target shared biology rather than individual diagnoses.
The study featured more than 500 authors, including three Pitt researchers: Mary Marazita, Distinguished Professor of Oral and Craniofacial Sciences, School of Dental Medicine, and professor of human genetics, School of Public Health; Ralph Tarter, professor of pharmaceutical sciences, School of Pharmacy; and Michael Vanyukov, professor of pharmaceutical sciences, School of Pharmacy, of psychiatry, School of Medicine, and of human genetics, School of Public Health. Marazita contributed nicotine dependence data, Tarter provided data about post-traumatic stress disorder and Vanyukov contributed substance use disorder data.
“The group was really interested in identifying groups of genes that influence multiple psychiatric disorders,” said Marazita. “It shows the promise of using these results to inform neurobiologically valid treatments and set targets for future development of treatments.”
The results found that many psychiatric conditions do share common genetic roots. Using statistics that examined all 14 disorders at once in the context of the entire genome, researchers identified five major factors: a factor underlying obsessive-compulsive disorder; a factor associated with schizophrenia and bipolar disorder; a neurodevelopmental factor connected to attention-deficit/hyperactivity disorder and autism spectrum disorder; an internalizing factor related to post-traumatic stress disorder, major depression, anxiety; and a substance use disorder factor including opioids, cannabis, alcohol and nicotine. Across all 14 disorders, these five factors accounted for 66% of the genetic variation—a statistic Marazita said is “remarkable.”
“You don’t usually identify that much of the variation,” Marazita said. “But having a large enough sample size and really good genomic resources was key.”
The analysis also uncovered 238 locations in the genome, known as loci, associated with one or more disorders. Most of these loci were pleiotropic, meaning they influence more than one psychiatric disorder, while a smaller number were disease-specific, affecting only a single disorder.
“There were relatively few disease-specific associations found,” said Marazita. “That’s not as many as you might have thought, given one million people with 14 disorders, but it’s a very interesting and important result that they did find some disease-specific ones.”
Many of the pleiotropic loci replicated and confirmed findings from past research, for example a well-known genetic “hot spot” on chromosome 11. In total, the study identified 101 regions of the genome where multiple disorders share genetic risk, with chromosome 11 being the most pleiotropic.
“The most well-replicated [hot spot] is on chromosome 11, where there’s a cluster of four genes,” Marazita said. “It is said to be a hot spot for a range of cognitive and behavior outcomes including intelligence, personality, substance use, sleep disorders and a number of other things.”
The pleiotropic loci were also linked to broader biological and neurobiological processes. Variants tied to the schizophrenia and bipolar factor, for example, were more active in the cells that send signals in the brain called excitatory neurons. Those in the internalizing pattern were linked to oligodendrocytes, cells that help speed neurological communication. Many pleiotropic variants were active early in life, even during prenatal brain development, suggesting that early biology influences risk before symptoms appear.
“With a lot of these disorders, you tend to think of them as adult onset,” said Marazita. “It’s interesting that these genes are active from a very early stage.” She said this result could allow researchers to seek targets for treatment that could be applied early in life.
For substance use disorders, the genetic pathways matched the specific substances—alcohol use disorder was associated with alcohol metabolism genes, while nicotine dependence was linked to nicotine receptors—with all the substance use disorders clustering in the same factor.
“You could have predicted some of these results in advance, but the importance of this study approach was that it is agnostic,” Marazita said. “They weren’t targeting any particular genes; they looked at all genes.”
While most of the study subjects were of European ancestry, Marazita said a smaller number were African or East Asian.
“A very important next step is to look at other ancestral backgrounds to see if the same degree of pleiotropy exists,” Marazita said. “You need huge collaborations. There’s no one group that can do it all, so it’s nice to see Pitt is part of these large collaborations that stand to make a big difference.”