Allelic association studies of genome wide association data can reveal errors in marker position assignments

Background: Genome wide association (GWA) studies provide the opportunity to develop new kinds of analysis. Analysing pairs of markers from separate regions might lead to the detection of allelic association which might indicate an interaction between nearby genes. Methods: 396,591 markers typed in 541 subjects were studied. 7.8*10(10) pairs of markers were screened and those showing initial evidence for allelic association were subjected to more thorough investigation along with 10 flanking markers on either side. Results: No evidence was detected for interaction. However 6 markers appeared to have an incorrect map position according to NCBI Build 35. One of these was corrected in Build 36 and 2 were dropped. The remaining 3 were left with map positions inconsistent with their allelic association relationships. Discussion: Although no interaction effects were detected the method was successful in identifying markers with probably incorrect map positions. Conclusion: The study of allelic association can supplement other methods for assigning markers to particular map positions. Analyses of this type may usefully be applied to data from future GWA studies.