Linkage detection by the Affected-Pedigree-Member method: What is really tested?
Abstract
We show that the null hypothesis in the Affected-Pedigree-Member (APM) method is composed of two sub-hypotheses: no linkage and correct marker allele distribution. As a result, the APM method is not robust to misspecification of marker allele frequencies. © 1993 Wiley-Liss, Inc.