For the last decade or more geneticists have been predicting that advances in molecular genetics are going to revolutionise our understanding of psychiatric disorders and human behavior. However, with a few exceptions, these expectations have yet to be fulfilled. As the century draws to a close and we contemplate the prospect of the complete sequence of the human genome it seems timely to consider the state of the field and to consider carefully how it might advance, the problems to be faced and the resources required. Molecular Psychiatry (2000) 5, 22-31.Keywords: molecular; epidemiology; linkage; linkage disequilibrium; phenotype; genomics Much of the optimism surrounding the application of molecular genetics to human disease has resulted from successes in applying these techniques to monogenic disorders. However, common illnesses pose much greater challenges for geneticists because, in the majority of cases, they result from the combined action of a number of different genes (each of which may result in only a modest increase or decrease in liability) as well as environmental influences; a witches' brew termed polygenic, multifactorial causation. Further complexities include the possibility of non-additive genetic effects, including gene-gene interactions (epistasis), and also potential gene-environment interactions. In psychiatry, genetic complexity is compounded by the nosological complexity of the phenotypes, which are best viewed as syndromes rather than diseases. It is rarely possible to validate psychiatric diagnoses on the basis of physical examination or laboratory tests or even to confirm them postmortem, and for many disorders we have little idea of pathogenic mechanisms.
Current state of progress
Genetic epidemiologyIn spite of these difficulties, genetic epidemiological studies based upon operational research diagnoses have shown that genes play a role in many of the syndromes defined by psychiatric nosology (Table 1) Heritability is the proportion of liability to a disorder accounted for by genetic effects. The heritability estimates are based on twin studies which employed DSM-III-R research diagnoses. NB: the heritabilities should be regarded only as approximations.both genes and environment contribute to most specific diagnostic categories, genetic epidemiologists have begun to explore the extent to which there is overlap in genetic and environmental risk factors between disorders; whether genes influence clinical heterogeneity within disorders and the relationship between genetic susceptibility to disorders and putative environmental risk factors. Patterns of familial co-aggregation have been important in helping us to extend diagnostic boundaries. Examples include the schizophrenic and autistic 'spectrum' disorders that are mild, often sub-clinical phenotypes associated with genetic liability to schizophrenia and autism respectively. 2 There is also evidence of a considerable overlap between the genetic contributions to depression and anxiety. 3 Such findings have important implications for...