The relation between sample sizes and statistical power is a constant issue in human molecular genetics. This relationship is not a simple one and is affected by several factors, such as number of statistical tests carried out and LD between markers. Idan Menashe et al presents in a recent paper in BMC Genetics a new freely-available program to calculate statistical power for distinct types of association studies. PGA is a program that is designed to give calculations of samples sizes required for specific statistical power values under different models, such as dominant, co-dominant or recessive effects for single SNPs or haplotypes, and for different study parameters (such as disease prevalences or allele frequencies). In addition, it is possible to obtain a minimum detectable relative risk for specific samples, which is useful for retrospective analysis of previously collected datasets.
PGA will be helpful, in addition to other available programs, such as the Power Genetic Calculator, for researchers designing and analyzing genetic association studies for a large number of human diseases and phenotypes. Author: Idan Menashe Biostatistics Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institute of Health, USA Email: menashei@mail.nih.gov Reference: Menashe I, Rosenberg PS, Chen BE. PGA: power calculator for case-control genetic association analyses. BMC Genet. 2008 May 13;9:36. Paper available online. The PGA program is available at PGA website.
Message posted by: Diego Forero
|