Dimensionality of Genomic Information and APY Algorithm
Ivan Pocrnić, Ignacy Misztal, Daniela A L Lourenco, et al.
Unpublished Poster, 2017
Abstract
ØSingle-step genomic best linear unbiased prediction (ssGBLUP) is a method for genomic evaluation that combines pedigree, data and genomic information in single analysis ØssGBLUPrequires inverse of genomic relationship matrix (G) which has a cubic cost if done directly ØGRM can be efficiently inverted by Algorithm for Proven and Young (APY) by recursion on a small number of core animals (Misztalet al. 2014) ØTheory by Misztal(2016): ØLimited dimensionality of genomic information ØLimited number of effective SNP markers (ESM) or independent chromosome segments (Me) ØE(Me)=4NeL where Ne is effective population size, and L is length of genome in Morgan (Stam, 1980)