This imputation algorithm uses LD to identify good predictors for each SNP, and then uses the high LD SNPs to identify K-Nearest Neighbors. The genotype is called with a weighted mode of the KNNs.
imputeLDKNNi(tasObj, highLDSSites = 30, knnTaxa = 10, maxDistance = 1e+07)
Number of sites in high LD to use in imputation.
Acceptable values are between
2000. Defaults to
Number of neighbors to use in imputation. Acceptable values
200. Defaults to
Maximum physical distance between sites to search for LD
(-1 for no distance cutoff - unlinked chromosomes will be tested).