Programs for determining parameters in complex genetic models for linkage analysis: GEN_PAR and EPI_PAR
✍ Scribed by Dr. Rosemarie Plaetke; William F. Byerley; Wolfgang Koehler; G. P. Vogler
- Publisher
- John Wiley and Sons
- Year
- 1991
- Tongue
- English
- Weight
- 168 KB
- Volume
- 8
- Category
- Article
- ISSN
- 0741-0395
No coin nor oath required. For personal study only.
✦ Synopsis
We want to report the availability of two programs, G E N J A R and EPLPAR. The programs determine certain epidemiological parameters that may assist in the choice of penetrances and allele frequencies of a disease-susceptibility gene for linkage analysis. The programs are written in PASCAL and are running on VAX computers.
Linkage analysis of diseases requires that the penetrances and the allele frequencies at the disease locus be defined. The magnitude of each of these parameters depends on the prevalence of the disease and the percentage of phenocopies among affected persons. For complex diseases, the mode of inheritance is usually not known. Neuman and Rice [ 19901 have proposed to simultaneously estimate penetrances and allele frequencies of the disease-susceptibility gene. Their approach considers the constraints that prevalence places on the genetic model. However, the genetic model is often a complicated one, because it may consist of more than one liability class. Additionally, the sample size may be too small for estimation of genetic parameters. To determine a genetic model for the analysis of schizophrenia, McGuffin et al. [ 19901 used the prevalence of the disease and assumed Hardy-Weinberg equilibrium at the disease locus. We adapted their approach in our programs G E N J A R and EPLPAR. Our approach for determining epidemiological and genetic parameters assumes that a disease locus has two alleles (d, D) and that its allele frequencies are in Hardy-Weinberg equilibrium.
G E N J A R . G E N J A R calculates upper limits for the penetrance of non-genetic cases and for the frequency l?rob(D) of the disease-susceptibility gene D. For these calculations we assume a certain prevalence P, frequency of phenocopies among the affected persons, and penetrances. The program works with an arbitrary number of liability classes. However, one must define the frequency of each liability class in the population.