Methods for measuring departures from
the Hardy-Weinberg equilibrium through
endogamy measures

Cibele Queiroz da-SILVA[1]

Talita ARMBORST[2]

§    ABSTRACT: One of the main measures of departure from the Hardy-Weinberg Equilibrium (HWE) is the endogamy coefficient, f. Such quantity measures heterozygotes deficit in the population. In this study three methods for estimating f are used: the Ayres and Balding's (AB) (1998) Bayesian method, and two classic ones, by Nei and Chesser (NC)(1983) and by Robertson and Hill (RH) (1984). We analyzed genotype frequency simulated data considering samples sizes 50, 200 and 1000 and 2, 6 and 15 alleles, with n=50 being the most realistic case. The methods were evaluated through the comparison of credibility intervals (via MCMC) and confidence intervals (calculated via Bootstrap). We found that the Ayres and Balding method is very efficient in producing estimates of f that lie in the parameter space. Moreover, the AB method has the advantage of allowing the calculation of allele frequencies. However, not only does the AB method demand too complex calculations, but also can produce very biased estimates of f, which are much larger than the ones observed for the other two methods.

§    KEYWORDS: Genetics; endogamy measures; MCMC; classic and Bayesian inference.



[1] Departamento de Estatística, Instituto de Ciências Exatas, Universidade de Brasília – UnB, CEP 70910-900, Brasília, Brasil, E-mail: cibeleqs@unb.br

[2] Inep/MEC - Esplanada dos Ministérios, Bloco L - Anexos II, sala 422 - CEP: 70047-900 - Brasília - DF, Brasil, E-mail: tarmborst@yahoo.com.br