Skip to contents

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.

Usage

imputeLDKNNi(tasObj, highLDSSites = 30, knnTaxa = 10, maxDistance = 1e+07)

Arguments

tasObj

an rTASSEL TasselGenotypePhenotype object.

highLDSSites

Number of sites in high LD to use in imputation. Acceptable values are between 2 and 2000. Defaults to 30.

knnTaxa

Number of neighbors to use in imputation. Acceptable values are between 2 and 200. Defaults to 10.

maxDistance

Maximum physical distance between sites to search for LD (-1 for no distance cutoff - unlinked chromosomes will be tested). Defaults to 10e6.