Premium
The haplotype runs test: The parent‐parent‐affected offspring trio design
Author(s) -
Lange Ethan M.,
Boehnke Michael
Publication year - 2004
Publication title -
genetic epidemiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.301
H-Index - 98
eISSN - 1098-2272
pISSN - 0741-0395
DOI - 10.1002/gepi.20010
Subject(s) - haplotype , transmission disequilibrium test , linkage disequilibrium , genetics , biology , allele , locus (genetics) , test statistic , statistics , statistical hypothesis testing , mathematics , gene
The increasing availability of maps of dense polymorphic markers makes use of haplotype data in family‐based association analyses an attractive alternative to single marker association tests. We describe a novel class of statistics designed to test for an association between marker haplotypes and a qualitative trait using the parent‐parent‐affected‐offspring trio design. Our haplotype runs test (HRT) is based on consecutive allele‐sharing between pairs of haplotypes. We assign weights according to the relative frequencies of the alleles for which the two haplotypes match. Herein, we compare the HRT to the maximum‐identity‐length‐contrast (MILC) statistic, the single‐locus transmission/disequilibrium test (TDT), and the generalized test of transmission disequilibrium for haplotype data, as implemented in the software TRANSMIT, using both simulated data and published haplotype data from the recessive disorder ataxia‐telangiectasia. Our simulation results suggest that the HRT outperforms the MILC and that the HRT provides comparable power to the TDT and TRANSMIT when the number of distinct founder haplotypes with a disease susceptibility allele is small but substantially outperforms the TDT and TRANSMIT when the number of distinct founder haplotypes with a disease susceptibility allele is even of modest size. © 2004 Wiley‐Liss, Inc.