Segregation analysis

We began the chapter by considering simple Mendelian inheritance. Segregation analysis was originally developed to test the hypothesis of a trait segregating in families according to either a dominant, recessive, or X-linked pattern. As such, the statistical procedures are straightforward, most commonly involving a c 2 goodness-of-fit test. The complications that arise largely result from differing methods of ascertainment. We have already mentioned this briefly in the earlier section discussing family studies. When a disorder is rare in a given population, such that the probability of ascertainment is low and no family contains more than one proband, this is called single ascertainment. The opposite is complete ascertainment when all probands are identified with certainty. In practice, with moderately common disorders there will be multiple-incomplete ascertainment with some families containing more than one proband (and hence being including in the sample more than once). The different methods of ascertainment affect the expected segregation ratios and hence they are important to specify. A detailed account is provided by Sham.(4)

The principles of segregation analysis have been extended to common disorders that do not show an obvious straightforward Mendelian pattern of transmission within pedigrees and where other evidence (for example less than 100 per cent concordance in monozygotic twins) indicate that there are environmental as well as genetic influences/5) The details of such methods are beyond our scope but in essence they can applied either to continuous traits or to present/absent traits (such as being affected or unaffected by disease) by assuming that there is a continuum of liability to the trait with only those beyond a certain threshold value manifesting it (see the discussion of thieshoid.modeis earlier in this chapter). Probably the most commonly applied and the most rigorous approach is embodied in the so called 'unified' model of Lalouel et al.(6) This allows the investigator to test how well a set of family data is explained by a combination of a major locus plus a multifactorial background and to compare this (as described earlier in the section on model fitting) with reduced models using a likelihood ratio test. The reduced models consist either of single gene alone or multifactorial transmission alone. If there is evidence for a single gene it is then possible to apply a further test of whether the transmission probabilities(7) conform to Mendelian expectation.

In practice, complex segregation analysis of this type has proved to be disappointing in psychiatric genetics. For example, in schizophrenia the findings have been inconclusive with the ability to distinguish between competing models being poor.(8) Furthermore, segregation analysis has produced misleading results such as suggesting an autosomal locus for the enzyme, platelet monoamine oxidase, which is now known to be encoded at two distinct X-linked loci.(9) Attempting to apply the approach to traits that are familial but not necessarily genetic can also be a pitfall for the unwary so that one study suggested that attendance at medical school among the relatives of medical students could best be explained by the segregation of a recessive gene. (!°.)

Hearing Aids Inside Out

Hearing Aids Inside Out

Have you recently experienced hearing loss? Most probably you need hearing aids, but don't know much about them. To learn everything you need to know about hearing aids, read the eBook, Hearing Aids Inside Out. The book comprises 113 pages of excellent content utterly free of technical jargon, written in simple language, and in a flowing style that can easily be read and understood by all.

Get My Free Ebook


Post a comment