Xref: utzoo sci.research:574 sci.bio:1723 sci.math.stat:517 bionet.molbio.evolution:51
Path: utzoo!attcan!uunet!yale!yalevm!BIO252
From: BIO252@YaleVM.YCC.Yale.Edu
Newsgroups: sci.research,sci.bio,sci.math.stat,bionet.molbio.evolution
Subject: statistics for testing spatial heterogeneity of mtDNA haplotype frequencies
Message-ID: <189@YaleVM.YCC.Yale.Edu>
Date: 10 Jan 89 20:47:46 GMT
Reply-To: BIO252@YaleVM.YCC.Yale.Edu
Organization: Yale University, New Haven, CT, USA
Lines: 24
Disclaimer: Author bears full responsibility for contents of this article

In a paper by DeSalle et al. (1987; Genetics 116:215-223) a method
is proposed to assess heterogeneity between populations using mtDNA
haplotypes using arsin, square-root transformed frequency of a given
haplotype in the following statistic:
 
 
 V = 4 x summation ni *((ai - mean a) squared)
 
 where ni = the number of individuals in the ith sample, ai = the
 
 
number of individuals in the ith pop. with the haplotype of interest,
mean a = sum ni x ai/N, N = total number of individuals in all pop.
and r = the number of populationns. Under the
null hypothesis of no geographical heterogeneity, V is distributed as a
Chi-Square with r-1 degrees of freedom.  The significance of Fst values
are tested with the V statistic.  In examples with two haplotypes,
alternate haplotypes give similar V values.  But where several haplotypes
occur, it appears that V values calculated for any one of them may
underestimate or over estimate between population heterogeneity.  Any
suggestions on how to alter this statistic to give an average of all
alleles?