SNPMace - A meta-analysis to estimate SNP effects across countries


  • Dr Jighly
  • Dr Benhajali Interbull Center
  • Dr Liu
  • Dr Goddard


The accuracy of genomic prediction could be improved by combining datasets across countries, but it is not always possible to combine the individual animal data. This project has tested a meta-analysis, called SNPMace, that mimics the combined analysis but requires only summary statistics, such as estimated SNP effects, from participating countries. The method uses the genetic correlation between a trait measured in different countries to produce country specific estimated SNP effects. We tested this method on data from 6 countries on the protein yield of Brown Swiss cattle and on the milk, fat and protein yields of Australian Holstein and Jersey cattle. In both cases the meta-analysis generated estimated breeding values that had a correlation with those obtained by analyzing the raw data in the range 0.99 to 1. The method is implemented in software called MetaGS which also converts data on a subset of SNPs to a common SNP set for analysis.