Skip to main content
Premium Trial:

Request an Annual Quote

Case-Case GWAS Approach Identifies Loci That Differ in Frequency Across Psychiatric Disorders

NEW YORK — Researchers have developed a new approach to uncover genetic loci that differ in allele frequencies across disorders, a method they applied to psychiatric disorders.

While a number of studies have emphasized genetic similarities between psychiatric disorders, two researchers from the Harvard T.H. Chan School of Public Health sought to use a new approach, callled case-case genome-wide association study, or CC-GWAS, to explore genetic differences between mood and psychotic disorders.

"Despite the large similarities between psychiatric disorders, it is clear that disorders also have very clear clinical distinctions," first author Wouter Peyrot, a visiting scientist at Harvard, said in an email. "To study genetic differences may help to improve differential diagnosis and provide more disorder-specific treatment in the future."

As they reported Monday in Nature Genetics, Peyrot and his colleague Alkes Price, a statistical geneticist at Harvard, developed CC-GWAS, which uses case-control GWAS summary statistics to detect differences in allele frequencies among cases from different studies. When they applied this approach to publicly available GWAS data of schizophrenia, bipolar disorder, major depressive disorder, and others, they uncovered nearly 200 case-case loci. Their analysis implicated genes belonging to the Krüppel-like family of transcription factors in schizophrenia, including one that had not previously been tied to the condition.

"This implies that CC-GWAS can identify novel disorder genes," Peyrot added.

Most approaches to determine differences in allele frequencies need individual-level data, the researchers noted. Their CC-GWAS approach, though, allows for such analysis using GWAS summary statistics by relying on a new genetic distance measure. 

When they applied the approach to publicly available summary statistics for studies of schizophrenia, bipolar disorder, and major depressive disorder, the analysis uncovered 121 loci across the three disorders. Of these, 21 were CC-GWAS-specific loci and eight had not been reported previously. 

The researchers also tied these CC-GWAS-specific loci to genes. For instance, based on summary-data-based Mendelian randomization data, the CC-GWAS-specific schizophrenia versus major depressive disorder locus pointed to seven genes, five of which encode protocadherin α proteins, which are involved in the development and function of cell-cell connections in the brains, and NDUFA2, which is linked to the early-onset progressive neurodegenerative disorder Leigh syndrome.

Two other CC-GWAS-specific loci — one identified in the schizophrenia versus bipolar analysis and the other in the schizophrenia versus major depressive disorder analysis — implicated the genes KLF16 and KLF6 in the Krüppel-like family of transcription factors. Both KLF16 and KLF6 are involved in neurite outgrowth and axon regeneration, and they might play a role in the synaptic pruning that is known to occur in schizophrenia.

The researchers further noted that while KLF6 had only recently been associated with schizophrenia, KLF16 had not previously been tied to the disorder, indicating that the CC-GWAS method could identify new disorder-associated genes.

They additionally analyzed data on five other psychiatric disorders — attention deficit-hyperactivity disorder, anorexia nervosa, autism spectrum disorder, obsessive-compulsive disorder, and Tourette's syndrome — in conjunction with the schizophrenia, bipolar disorder, and major depressive disorder datasets and uncovered 72 CC-GWAS-specific loci. 

To Peyrot, this also underscored the advantage of CC-GWAS, as it was able to conduct 28 different case-case comparisons across as many as eight different disorders. He noted that an important previous study was only able to compare two disorders, schizophrenia and bipolar disorder, because of the need to match individual-level data.

He and his colleagues are now developing additional analytical methods to study psychiatric and other disorders, though other researchers are applying CC-GWAS to other datasets, according to Peyrot.