1. Genetic databases are an expanding and readily accessible repository of information on the mapping and sequencing of the human genome, and that of other model organisms. The integration and application of this information to neuropsychiatric disease is illustrated using neuroendocrine and neuropharmacologic data, computerized and other genetic databases. 2. This computer-assisted integrated approach to knowledge structures permits the rapid generation of hypotheses, the prompt identification of candidate gene loci, an explanation for genetic heterogeneity, and strategies for the use of potential linked markers. 3. Results using this integrated search strategy demonstrate that over 30 candidate loci for neuropsychiatric disease have currently been mapped in man (spread over 14 chromosomes in the human genome), and that at least 6 homologous loci have been mapped in mouse. 4. Using a metabolic pathway approach, it can be shown that the best current candidate gene locus for a subtype of schizophrenia located on chromosome 5q11-13 (HGML10 # SCZD1 and OMIM #181510) is in the serotonergic pathway, i.e. a receptor for 5-hydroxytryptamine (subtype 1A; HGML10 #HTR1A) which also maps in the same chromosomal region. 5. Parallels are suggested between inborn errors of metabolic pathways in the somatic endocrine system (using insulin-dependent diabetes mellitus as a paradigm) and the neurotransmitter and hormonal systems within the brain. 6. A subset of neuropsychiatric disorders may thus be viewed as inborn errors of cerebral metabolic pathways primarily affecting the biogenic amine pathways.