In a study of more than 100,000 people, researchers at the Icahn School of Medicine at Mount Sinai used a novel machine-learning method to identify 413 genetic associations with schizophrenia across 13 regions of the brain. By examining gene expression at the tissue level, researchers identified new genes associated with schizophrenia and were also able to pinpoint areas of the brain in which abnormal expression might occur.
Although it affects less than 2% of the population worldwide, schizophrenia is one of the main causes of disability. The disease has a major public health and socioeconomic impact, primarily because of hospital readmission rates and treatment costs.
The researchers used genome-wide association study findings together with transcriptomic imputation to identify schizophrenia-associated disease with tissue-level resolution. Genome-wide association studies examine differences at various points in a genetic code to see whether a variation occurs more frequently in those with a particular trait, such as schizophrenia. Transcriptomic imputation, a novel machine-learning technique, enables researchers to test associations between disease and gene expression in otherwise unreachable tissues, such as brain tissue.
Studying 40,299 people with schizophrenia and 62,264 matched controls, the researchers discovered that genes associated with schizophrenia are expressed throughout development–some during early stages of pregnancy, and others during adolescence or adulthood. They also discovered that different brain regions confer different risks for schizophrenia, and most associations stem from the dorsolateral prefrontal cortex.
Source: MedicalXpress, March 25, 2019