Rapid Responses to:

LETTERS: J H Edwards Unifactorial models are not appropriate for multifactorial disease BMJ 1999; 318: 1353b [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Does Genealogy Matter for Complex (Multifactorial) Genetic Disease?

Augustine Kong   (30 June 1999)

Does Genealogy Matter for Complex (Multifactorial) Genetic Disease? 30 June 1999
Augustine Kong Send response to journal: Re: Does Genealogy Matter for Complex (Multifactorial) Genetic Disease?	Professor Jonathan Edwards published a letter in the BMJ, the aim of which we have some difficulties understanding. His first paragraph suggested that it was not about ethical or political issues, but the last paragraph did. His collaborations with some of our local critics, who have been doing research on the genetics of schizophrenia may explain this. However, what is clear is that Professor Edwards takes the rather narrow view that genealogy information is only useful for diseases which can be fitted with "unifactorial models". We disagree. Data are the fuel of all scientific investigations, and in the case of genetic studies, it is intuitively obvious that phenotypic, genetic, and genealogical data are all relevant. All methods for finding susceptibility genes with anonymous markers (even those which do not assume "unifactorial" etiology) involve searching for genomic regions which patients share. Apart from power considerations, one aspect that distinguishes the various approaches is resolution. For linkage/family studies a low density of markers is necessary to find sharing, but it results in low resolution. Association techniques.studies that focus on "unrelated", or very distantly related affected individuals can provide very fine resolution, but a very high density of markers is required to capture the sharing. Given our genealogy, at any stage of the gene mapping effort, we have the flexibility to study affected relatives separated by any distance desirable and matching it with the marker density required for that distance. We can perform studies which are half way between the traditional family studies and association studies and avoid having to jump directly from one extreme to the other. Our extensive genealogy database allows us to fractionate the patient material, perhaps making the genetic basis for the disease in this subgroup even simpler. The proposed Icelandic Healthcare Database (www.database.is) would make anonymous and combine such data for a large proportion of the population (those who give active informed consent for the genetic data) allowing one to ask fundamental questions about how interactions among genes or between genes and the environment, cause disease. Mapping genes for common disorders is not easy and combining extensive genealogy of an isolated population with powerful genome sharing methods represent new approaches to human genetics. But this may be irrelevant to Professor Edwards who seems to have contempt for most "state of the art" techniques. Augustine Kong PhD Director of Statistical Genetics Jeff Gulcher MD PhD Vice President of Research and Development Kari Stefansson MD PhD Chief Executive Officer Decode Genetics Lynghals 1 Reykjavik Iceland 110