Perturbation analysis: a simple method for filtering SNPs with erroneous genotyping in genome-wide association studies.


Teo, YY; Small, KS; Clark, TG; Kwiatkowski, DP; (2008) Perturbation analysis: a simple method for filtering SNPs with erroneous genotyping in genome-wide association studies. Annals of human genetics, 72 (Pt 3). pp. 368-74. ISSN 0003-4800 DOI: https://doi.org/10.1111/j.1469-1809.2007.00422.x

Full text not available from this repository.

Abstract

We introduce a simple and yet scientifically objective criterion for identifying SNPs with genotyping errors due to poor clustering. This yields a metric for assessing the stability of the assigned genotypes by evaluating the extent of discordance between the calls made with the unperturbed and perturbed intensities. The efficacy of the metric is evaluated by: (1) estimating the extent of over-dispersion of the Hardy-Weinberg equilibrium chi-square test statistics; (2) an interim case-control study, where we investigated the efficacy of the introduced metric and standard quality control filters in reducing the number of SNPs with evidence of phenotypic association which are attributed to genotyping errors; (3) investigating the call and concordance rates of SNPs identified by perturbation analysis which have been genotyped on both Affymetrix and Illumina platforms. Removing SNPs identified by the extent of discordance can reduce the degree of over-dispersion of the HWE test statistic. Sensible use of perturbation analysis in an association study can correctly identify SNPs with problematic genotyping, reducing the number required for visual inspection. SNPs identified by perturbation analysis had lower call and concordance rates, and removal of these SNPs significantly improved the performance for the remaining SNPs.

Item Type: Article
Faculty and Department: Faculty of Epidemiology and Population Health > Dept of Infectious Disease Epidemiology
Faculty of Infectious and Tropical Diseases > Dept of Pathogen Molecular Biology
PubMed ID: 18261185
Web of Science ID: 254927100009
URI: http://researchonline.lshtm.ac.uk/id/eprint/4450

Statistics


Download activity - last 12 months
Downloads since deposit
0Downloads
270Hits
Accesses by country - last 12 months
Accesses by referrer - last 12 months
Impact and interest
Additional statistics for this record are available via IRStats2

Actions (login required)

Edit Item Edit Item