Haplotype Tagging Single Nucleotide Polymorphisms and Association StudiesThompson D. · Stram D. · Goldgar D. · Witte J.S.
aUnit of Genetic Cancer Epidemiology, International Agency for Cancer Research, Lyon, France; bCR-UK Genetic Epidemiology Unit, University of Cambridge, UK; cDepartment of Preventive Medicine, University of Southern California, Los Angeles, Calif., and dDepartment of Epidemiology and Biostatistics, University of California, San Francisco, Calif., USA
Objectives: Discrete blocks of low haplotype diversity exist within the human genome. The non-redundant subset of ‘haplotype tagging’ single nucleotide polymorphisms (htSNPs) in such blocks can distinguish a majority of the haplotypes. Several approaches have been proposed to determine htSNPs, ranging from visual inspection to formal analytic procedures. Optimal htSNPs can be estimated using a small subgroup of an association study population that have been genotyped for a dense SNP map, and it is just these htSNPs that are genotyped in the remainder of the samples. We investigated by simulation how the size of the subsample affects the power of association studies, and what type of subjects it should include. Methods: We used the program tagSNPs [Stram et al., Hum Hered 2003;55:27–36], which selects htSNPs to minimize the uncertainty in predicting common haplotypes for individuals with unphased genotype data. Results: On average, 27% of the SNPs were designated as htSNPs. Genotyping as few as 25 unphased individuals to select the htSNPs did not appear to reduce the power of an association study, as compared with using all SNPs. For the disease models considered, selecting htSNPs based on cases, controls, or a mixture of both gave similar results. Conclusions: These results suggest that the genotyping effort in an association study can be substantially reduced with little loss of power by identifying htSNPs in a small subsample of individuals.
John S. Witte
Department of Epidemiology and Biostatistics
500 Parnassus Ave, MU – 420 West
San Francisco, CA 94143-0560 (USA)
Tel. +1 415 476 1374, Fax +1 415 476 6014, E-Mail email@example.com
Received: April 30, 2003
Accepted after revision: July 8, 2003
Number of Print Pages : 8
Number of Figures : 1, Number of Tables : 4, Number of References : 24
Human Heredity (International Journal of Human and Medical Genetics)
Founded 1950 as Acta Genetica et Statistica Medica by Gunnar Dahlberg; Continued by M. Hauge (1965–1983)
Vol. 56, No. 1-3, Year 2003 (Cover Date: Released November 2003)
Journal Editor: J. Ott, New York, N.Y.
ISSN: 0001–5652 (print), 1423–0062 (Online)
For additional information: http://www.karger.ch/journals/hhe