Comparison of genotyping using pooled DNA samples (allelotyping) and individual genotyping using the affymetrix genome-wide human SNP array 6.0

Abstract

BackgroundGenome-wide association studies (GWAS) using array-based genotyping technology are widely used to identify genetic loci associated with complex diseases or other phenotypes. The costs of GWAS projects based on individual genotyping are still comparatively high and increase with the size of study populations. Genotyping using pooled DNA samples, as also being referred as to allelotyping approach, offers an alternative at affordable costs. In the present study, data from 100 DNA samples individually genotyped with the Affymetrix Genome-Wide Human SNP Array 6.0 were used to estimate the error of the pooling approach by comparing the results with those obtained using the same array type but DNA pools each composed of 50 of the same samples. Newly developed and established methods for signal intensity correction were applied. Furthermore, the relative allele intensity signals (RAS) obtained by allelotyping were compared to the corresponding values derived from individual genotyping. Similarly, differences in RAS values between pools were determined and compared.ResultsRegardless of the intensity correction method applied, the pooling-specific error of the pool intensity values was larger for single pools than for the comparison of the intensity values of two pools, which reflects the scenario of a case–control study. Using 50 pooled samples and analyzing 10,000 SNPs with a minor allele frequency of >1% and applying the best correction method for the corresponding type of comparison, the 90% quantile (median) of the pooling-specific absolute error of the RAS values for single sub-pools and the SNP-specific difference in allele frequency comparing two pools was 0.064 (0.026) and 0.056 (0.021), respectively.ConclusionsCorrection of the RAS values reduced the error of the RAS values when analyzing single pool intensities. We developed a new correction method with high accuracy but low computational costs. Correction of RAS, however, only marginally reduced the error of true differences between two sample groups and those obtained by allelotyping. Exclusion of SNPs with a minor allele frequency of ≤1% notably reduced the pooling-specific error. Our findings allow for improving the estimation of the pooling-specific error and may help in designing allelotyping studies using the Affymetrix Genome-Wide Human SNP Array 6.0.

DNA methylation levels are highly correlated between pooled samples and averaged values when analysed using the Infinium HumanMethylation450 BeadChip array

“…Population stratification needs to be excluded. Furthermore, the error rate tends to be higher in pooled samples compared to individual ones [ 21 ]. It is also important in the study design for EWAS with pooled samples to take into account the sample size needed to compute DMCs with confidence.…”

Derived Polymorphic Amplified Cleaved Sequence (dPACS): A Novel PCR-RFLP Procedure for Detecting Known Single Nucleotide and Deletion–Insertion Polymorphisms

“…Importantly, SNP/DIP detection methods have dramatically improved over time, especially with regard to sample throughput and accuracy. It is now possible to analyse hundreds of thousands of SNPs in a few samples with DNA chip-based methodologies [40]. When a large number of individuals require genotyping for a few SNPs/DIPs, the probe-based Taqman and the more recent Kompetitive Allele Specific PCR (KASP) technology are preferred [41,42].…”

Pooled genome wide association detects association upstream of FCRL3 with Graves’ disease

“…A pooled DNA study has also identified a new susceptibility locus, HGF , in keratoconus (a progressive corneal degeneration) and additional novel candidate genes related to platelet reactivity in diabetic patients [34–36]. The study design must be carefully considered to include multiple pools to minimise pooling errors and an appropriate array platform to extract maximal available information from the pooled DNA [37, 38]. …”