Genotyping efficiency, or call rate, is an issue which will be discussed in greater depth in the Marker Quality section below. A large proportion of SNP assays failing on an individual DNA sample may be indicative of a poor quality DNA sample, which could lead to aberrant genotype calling. Samples with low genotyping efficiency, or call rate, should be eliminated from further analysis. A recommended threshold is 98-99% efficiency, after first removing markers which have a low genotype call rate across samples. The suggested 98-99% threshold is an approximate threshold – the exact threshold may vary from study to study depending on the genotyping platform used, quality of the DNA samples used, and the variability in human and equipment error in genotyping. The threshold should be determined based on a goal whereby a balance minimizing the number of samples dropped and maximizing genotyping efficiency is attained. Figure 6 shows the proportion of samples (red and blue lines) or SNPs (green line) remaining at different call rate thresholds. Genotyping efficiency can be checked using the --missing option in PLINK. This will
