Email updates

Keep up to date with the latest news and content from BMC Genetics and BioMed Central.

This article is part of the supplement: Genetic Analysis Workshop 14: Microsatellite and single-nucleotide polymorphism

Open Access Proceedings

Linkage analysis of complex diseases using microsatellites and single-nucleotide polymorphisms: application to alcoholism

Jérémie Nsengimana12*, Hélène Renard1 and David Goldgar1

Author Affiliations

1 Genetic Epidemiology Group, International Agency for Research on Cancer, World Health Organization, 150 Cours Albert Thomas, 69008 Lyon, France

2 Saint James's University Hospital, Genetic Epidemiology Division, Cancer Research UK Clinical Centre in Leeds, Cancer Genetics Building, Beckett Street, Leeds LS9 7TF, UK

For all author emails, please log on.

BMC Genetics 2005, 6(Suppl 1):S10  doi:10.1186/1471-2156-6-S1-S10

Published: 30 December 2005


The efficacy of linkage studies using microsatellites and single-nucleotide polymorphisms (SNPs) was evaluated. Analyzed data were supplied by the Collaborative Study on the Genetics of Alcoholism (COGA). Alcoholism was analyzed together with a simulated trait caused by a gene of known position, through a nonparametric linkage test (NPL). For the alcoholism trait, four densities of SNPs (1 SNP per 0.2 cM, 0.5 cM, 1 cM and 2 cM) showed higher peaks of NPL z scores and smaller significant p-values than the usual 10-cM density of microsatellites. However, the two highest densities of SNPs had unstable z score signals, and therefore were difficult to interpret. Analyzing a simulated trait with the same markers in the same pedigrees, we confirmed the higher power of all four densities of SNPs compared to the 10-cM microsatellites panel, although the existence of other confounding peaks was confirmed for maps that are denser than 1 SNP/cM. We further showed that estimating the gene position using SNPs is far less biased than using the usual panel of microsatellites (biases of 0–2 cM for SNPs vs. 8.9 cM for microsatellites). We conclude that using dense maps of SNPs in linkage analysis is more powerful and less biased than using the 10-cM maps of microsatellites. However, linkage signals can be unstable and difficult to interpret when several SNPs are genotyped per centimorgan. The power and accuracy of 1 SNP/cM or 1 SNP/2 cM may be sufficient in a genome-wide linkage scan while denser maps may be most useful in fine-gene mapping studies exploiting linkage disequilibrium.