The Genomics of Psychiatric Illnesses: Moving Forward

Neuropsychiatric disorders are genetically complex, multifactorial disorders. The growing set of risk variants identified in large-scale genomewide association studies of psychiatric disorders and including 100,000–200,000 participants indicates a complex pattern, where many risk variants each contribute a small incremental risk (1). It is increasingly unlikely that a major single gene predisposes or directly causes any major psychiatric disorder but rather, many genes make minor contributions to disease risk. In addition, risk also seems multifactorial, where genes alone do not cause disease; instead, an interplay between genes and environmental factors is most likely at the root of causation. As we define the polygenic risk profiles of our patients, we need to define the genomic profiles that are protective and promote resilience to risk.

Despite our growing understanding of the complexity of neuropsychiatric genomics, the importance of heritable risk remains clear. Historically, twin and family studies examining the pattern of disease occurrence have demonstrated shared genetic risk factors. For example, twin studies examine concordance (both twins ill or not ill) and discordance (one twin with illness, the other without) of illness. Because monozygotic (MZ) twins share almost all their genes and dizygotic (DZ) only half their genes on average (like any pair of siblings), rates of co-occurrence of disease in MZ twins versus DZ twins can be used to estimate genetic heritability. Typically, these studies have found MZ concordance rates around .50 and DZ rates around .20 (2). Family studies where multiple members are affected will continue to be valuable populations for study.

As we continue to explore psychiatric phenotypes for genomic analysis, it may be particularly important to look at specific symptoms, symptom severity, comorbidities, onset, and population variations such as sex, race/ethnicity, and the unknown contributions of unspecified environmental factors. Bigger samples of study participants for any given psychiatric phenotype are needed. Using a modern research design in which participants can join the research via secure Internet avenues and follow-up involves face-to-face contact with research clinicians, we hope to reach affected individuals who may and may not come to academic research centers for care. As clinicians as well as researchers, we need to engage patients, family members, and clinicians to encourage participation (3).