Neuropsychiatric disorders such as schizophrenia demonstrate a high degree of heritability. Reference Sullivan, Kendler and Neale1,Reference Sullivan, Daly and O'Donovan2 Understanding the genetic component of these disorders has driven decades of rigorous research in psychiatric genetics. Reference Williams, Owen and O'Donovan3,Reference Moore, Kelleher and Corvin4 Building on the success of earlier techniques such as linkage studies, recent genome-wide association studies (GWAS) Reference Purcell, Wray, Stone, Visscher, O'Donovan and Sullivan5-7 have begun identifying risk variants that may give us important insight into disease mechanisms. There remains, however, significant ‘missing heritability’ Reference Eichler, Flint, Gibson, Kong, Leal and Moore8 in these complex diseases. Findings from GWAS have also demonstrated that many risk alleles do not necessarily code for amino acid changes in proteins, indicating a more regulatory function. Reference Stefansson, Ophoff, Steinberg, Andreassen, Cichon and Rujescu9 The role of gene-gene and gene-environment interactions in the expression of disease phenotype and timing of illness onset have been suggested, but with the exception of some epigenetic phenomena Reference Tsankova, Renthal, Kumar and Nestler10 we understand relatively little about these mechanisms.
The significance of non-coding regions
The Encyclopedia of DNA Elements (ENCODE) project has assigned biochemical functions to 80% of the genome and in doing so has laid to rest the idea that the vast areas of non-protein-coding DNA discovered by the Human Genome Project is ‘junk DNA’. Reference Bernstein, Birney, Dunham, Green, Gunter and Snyder11 It has emerged that non-coding regions are involved in a large number of regulatory processes including gene-gene regulation, gene-protein interaction and the transcription of non-translated RNA.
By highlighting the importance of non-coding functional DNA, ENCODE will allow researchers to re-evaluate the significance of existing psychiatric GWAS findings. We already know that certain regulatory sites harbour GWAS variants that are strongly correlated with the promoter regions of genes associated with schizophrenia. Reference Maurano, Humbert, Rynes, Thurman, Haugen and Wang12 Regulatory elements close to neuronal growth genes are highly preserved in the human lineage, indicating previously unsuspected functional importance. Reference Ward and Kellis13
Mechanisms for gene regulation, and gene-gene and gene-environment interaction
ENCODE found that 95% of the genome was within 8 kb of a protein-gene interaction. Reference Bernstein, Birney, Dunham, Green, Gunter and Snyder11 These findings may frame new hypotheses for mechanisms of gene-environment interactions, why disorders associated with similar genetic risk Reference Purcell, Wray, Stone, Visscher, O'Donovan and Sullivan5 present with varied phenotypes and why so many disorders have onset at specific points in life.
Protein coding accounts for only 2% of the genome and a further 75% can be transcribed into non-translated RNA, with a likely regulatory role in at least some cell types. Reference Djebali, Davis, Merkel, Dobin, Lassmann and Mortazavi14 The fundamental concept of what a gene actually is may require some rethinking Reference Ecker, Bickmore, Barroso, Pritchard, Gilad and Segal15 and the role of non-translated segments of RNA in disease will be a new focus for medical genetics.
Finally, ENCODE has demonstrated how distant genes interact to affect each other, Reference Dostie, Richmond, Arnaout, Selzer, Lee and Honan16 revealing what has been termed a ‘3D puzzle’ of gene regulation. Reference Pennisi17 This may explain how multiple distant and unrelated genes interact to produce complex clinical phenotypes.
Conclusions
Psychiatry has good cause to feel excitement over ENCODE. Following decades of research into psychiatric genetics, few conclusive findings have been translated into clinical practice. Alongside the recent successes of modern psychiatric genetics, ENCODE offers a much greater appreciation of the complexity of genomic biology. In this new scientific landscape psychiatrists should find some important answers to long-asked questions.
eLetters
No eLetters have been published for this article.