NEW YORK (GenomeWeb) – A large new case-control study has confirmed the over-representation of copy number variants in individuals with schizophrenia, and has expanded the collection of rare CNVs and related pathways associated with the condition.
Members of the CNV and Schizophrenia Working Groups of the Psychiatric Genomics Consortium did array-based CNV profiling on tens of thousands of individuals with or without schizophrenia. Their analyses, reported today in Nature Genetics, unearthed schizophrenia-associated CNVs and pathways that had previously been linked to the condition, and highlighted new variants, genes, and pathways not picked up in prior studies with smaller cohorts.
"Our data suggest that the undiscovered novel risk loci that can be detected with current genotyping platforms lie at the ultra-rare end of the frequency spectrum and still larger samples will be needed to identify them at convincing levels of statistical evidence," corresponding and co-senior author Jonathan Sebat, a psychiatry, cellular, and molecular medicine researcher at the University of California at San Diego, and his colleagues wrote.
While CNVs have been implicated in schizophrenia in the past, the study's authors noted that the current analysis represents "the largest genome-wide analysis of CNVs for any psychiatric disorder to date."
The team brought together data from dozens of past studies for 21,094 individuals with schizophrenia and 20,227 unaffected control individuals profiled with Illumina and Affymetrix array, and used its analytical pipeline to consistently call CNVs across the samples. Through a series of analyses focused on genome-wide CNV patterns, CNV-affected genes or pathways, and CNV breakpoints, the group then searched for patterns that were over-represented in the schizophrenia cases compared with controls.
After recognizing and removing nine CNVs with known links to schizophrenia, the researchers narrowed in on eight new CNV loci with genome-wide significant schizophrenia associations, including six copy number losses and two gains. Across the complete schizophrenia cohort, however, these CNVs were found in just 1.4 percent of cases, consistent with the relative rarity of each associated variant. Eight more CNV sites showed weaker ties to the condition.
The researchers' analyses also suggested there may be CNVs that protect against schizophrenia, though they cautioned that variants associated with reduced schizophrenia risk may not necessarily protect against other forms of neurodevelopmental disease.
"It is probable that some of the undiscovered rare alleles affecting risk for [schizophrenia] confer protection," they wrote, "but larger sample sizes are needed to determine this unequivocally."
When they specifically considered three-dozen gene sets related to synaptic functions, neurodevelopmental processes, and other neurological features, meanwhile, the investigators identified 15 genes sets that were more frequently affected by losses in the schizophrenia cases compared with controls. Gains were over-represented in the schizophrenia group in four other gene sets.
Likewise, the team's network pointed to an over-representation of deletions in the suspicious pathways, while its breakpoint analysis led to sites that seemed particularly prone to deletions or other CNVs.
The researchers noted that still-larger studies will be needed to address the role that rarer CNVs or CNVs with smaller effect sizes might play in schizophrenia, since the current study focused on CNVs with large effect sizes that are present in more than 0.1 percent of individuals with schizophrenia.