The genotype measurement is a discrete variable which can take four values: homozygote for the major allele, homozygote for the minor allele, heterozygote and unknown: g s, i)∈{ maj, min, het, unk}.
We investigated whether the low variance explained in KORA-F3 might have been due to a problem of genotype measurement error due to the lower average imputation quality (median Rsq = 0.68 versus Rsq = 0.95 and 0.99 in MICROS and popgen, respectively).
Our data include incomplete genotype measurements in the studied founders, denoted by GF, incomplete genotype measurements in the offspring, denoted by GO, and the ascertainment status of each studied individual, represented by a vector D pointing at individuals with an ascertainment proband status.
We compared the microarray measurements between genotype and DNA methylation of the probes selected in common and found that the microarray DNA methylation measurements were always less than the genotype measurements.
We determined that 28 people did not have any genotype measurements and had to exclude them from the analysis, leaving a sample of 248 workers.
In the preprocessing step, the data are filtered and uncertain genotype measurements, as well as individuals and SNPs that do not fit quality requirements are discarded.
We previously reported such a method, called Parental Support, which uses single-nucleotide polymorphism (SNP) microarray measurements from maternal and paternal DNA to improve the fidelity of noisy embryonic DNA genotype measurements.
Given the very large number of phenotype and genotype measurements taken the results of all the analyses undertaken are presented in additional chromosome figures with results from each of the 26 chromosomes presented in a series of 21 graphs.
However, when used in combination with SE genotyping, measurement of aCCP appears to distinguish a subset of patients who may differ with respect to response to therapy and outcome.
Non-differential misclassification bias from genotyping measurements is possible although it is unlikely in the absence of Hardy-Weinberg disequilibrium and complete concordance for repeated samples.
We applied experimental designs to minimize technical bias, described in Methods, and plots of the genotyping measurements (P1[B]/P1[A] ratio, mean LRR, and mean BAF) do not provide any evidence for bias between cases and controls (Supp. Figs. S1 and S6).
