| Phylogenetic methodologies, traditionally used in evolutionary biology, have many other potentially useful applications to the genetics of complex disease. The first of which relates to the concept of a continuum of psychosis which suggests that the current diagnostic paradigm in psychiatry creates artificial boundaries where instead the reality may be one of continual overlap of patients' symptoms as well as the true etiological origin of these symptoms. Such artificial boundaries create a heterogeneous pool of patients in any one diagnostic group, which is particularly problematic in genetic studies of psychiatric phenomena where phenotypic homogeneity is essential for the identification of genetic loci contributing to a disease state. Here we hypothesize that if psychiatric patients are grouped together based on behavioral symptoms using a character based clustering approach, without regard to clinical diagnosis (i.e. schizophrenic, bipolar), subgroups of patients whose symptoms share the same genetic origin would be identified. The goal is to improve the granularity of the psychiatric diagnoses by identifying clusters of behavioral symptom profiles that might share common genetic factors. Upon analysis of a psychiatric dataset we successfully separated and amplified two linkage peaks linked to the diagnostic category of major depression in our dataset. One peak is linked to a group of depressed individuals with specific sleep disturbance symptoms on chromosome 2q13. The other peak is linked to a mutually exclusive group of individuals with symptoms of depression but almost none of the specific sleep disturbance issues seen in the other group on chromosome 2q35. In addition we identified a subset of schizophrenics defined by specific language disturbances linked to a genetic marker on chromosome 2p25.1 and a group of patients with a phenotype intermediate between those of schizophrenia and schizoaffective disorder on chromosome 2p21.;The next potential application involves testing whether reconstructing the genetic history of a population is an effective way of isolating which genetic mutations are associated with a disease. This methodology has effectively identified variants of genes associated with a disease state under a number of simulated scenarios and has been successful in identifying genetic variants associated with age-related macular degeneration. |
