Consequently, DIST is useful for investigators who need fast and fairly accurate access to imputation-based P-values but (i) do not have access to subject-level genotypes, (ii) do not want to go through the laborious process of imputing subject-level genotypes or (iii) have association statistics coming from (large) pedigree data.
When compared with genotype imputation methods, DIST (i) does not need access to subject-level genotypes, (ii) provides comparable imputation accuracy while substantially shortening the running time and (iii) can be readily applied to family association statistics.
Unfortunately, the commonly used genotype imputation methods need access to genetic data, which, unlike summary statistics, are not always available.
Unlike freely available summary statistics, there is a limited (or, at least, not timely) access to genotypic data that is required by genotype imputation methods.
Missing cost data, for example, from patients who did not consent to access to their MBS, PBS, and public hospital data will be imputed using the Multiple Imputation using Chained Equations (MICE) approach.
A drawback of standard imputation is that it, in principal, requires access to an appropriate reference sample; results can be highly sensitive to the reference panel used (as seen from the disparity between our imputation results around the MalariaGEN causal SNP, rs334, using 1000 Genomes as a reference panel, compared to those obtained by Band et al. [ 2013] using HapMap3 as a reference panel).
Sparse genotyping combined with imputation may be a viable alternative, but requires access to high-density marker data and genetic maps, which may not be available in most aquaculture species in the near future.
In addition to imputation, we also performed joint-testing in the simulated data because we had access to the individual genotypes and thus they could compute the true SNP correlation matrix.
Limited access to doctors?
Limit access to resources.
