NAMHC Genomics Workgroup: Research Recommendations Summary

The National Institute of Mental Health (NIMH) convened a workgroup to advise the National Advisory Mental Health Council on addressing the gap in knowledge between gene discovery and understanding the potential importance of genomic findings in the causation, risk, trajectory, or resilience to psychiatric disorders. The NIMH Council Workgroup on Genomics recognized the challenges and opportunities in psychiatric genomics, and provided guidance in many areas relevant to genomics research. The following is a high-level summary of the workgroup’s recommendations for research moving forward. For additional information, please see the full workgroup report, Report of the National Advisory Mental Health Council Workgroup on Genomics, as well as a discussion of these recommendations in a Director’s Message, Towards a Genomic Psychiatry: Recommendations of the Genomics Workgroup of the NAMHC.

• Employ appropriate statistical methods and adopt rigorous significance standards for disease association. Strict standards of statistical rigor are necessary to interpret both discovery of common or rare genetic variation in clinical studies and to justify biological follow-up experiments to understand the functional roles of gene variants.
• Abandon candidate gene approaches in favor of well-powered, unbiased genetic association studies for gene discovery. In this context, a ‘candidate gene’ is defined as a gene selected for study based on prior biological hypotheses, rather than an unbiased, genome-wide approach.
• Explore all types of genetic variation for disease association. To understand the full spectrum of genetic risk, it is critical to study both common and rare gene variants, single base pair changes, as well as larger and more complex forms of variation.
• Expand genetic association efforts beyond the DSM. Heritability is shared across psychiatric disorders, as well as across related dimensional traits. Understanding the relevant biological pathways requires analysis not only of traditional disease categories, but also of functional domains (e.g., valence, cognition) affected by mental illnesses.
• Capture genetic and phenotypic variation across diverse human populations. Studies of diverse populations can identify novel disease-associated loci that may be unique to a particular population, as well as contribute to fine mapping of previously identified loci.
• Develop and share research resources. Databases, brain, DNA and cell repostiories, bioinformatics resources and statistical methods should be collected, stored, and made available for distribution to the research community.
• Require robust, genome-wide significance in selecting genes and gene variants for further study. After the principles of statistical rigor are met, selection criteria for putative variants for biological studies should be based on the type of variant. For example, identifying causal common variants requires fine mapping of genome-wide association studies (GWAS) loci. Rare variants must meet similar rigorous statistical thresholds. Importantly, follow-up studies of genes identified through candidate gene approaches should no longer be of interest.
• Carefully choose experimental systems for biological studies. There is a recognized need to use appropriate experimental systems – including model animals, in vitro human or animal cell or tissue systems, and computer models –  to investigate gene function and the biological effects of disease-associated variants. However, such studies should be understood as basic science investigations rather than as the generation of “disease models” representing human etiology or pathophysiology. The choice of experimental system will depend on the biological question and consideration of factors such as technological feasibility, evolutionary conservation, and genetic background.