Methods and Strategies to Impute Missing Genotypes for Improving Genomic Prediction

Publikation: Bog/antologi/afhandling/rapportPh.d.-afhandlingForskning

  • Peipei Ma, Danmark
Genomic prediction has been widely used in dairy cattle breeding. Genotype imputation is a key procedure to efficently utilize marker data from different chips and obtain high density marker data with minimizing cost. This thesis investigated methods and strategies to genotype imputation for improving genomic prediction. The results indicate the IMPUTE2 and Beagle are accurate imputation methods, while Fimpute is a good alternative for routine imputation with large data set. Genotypes of non-genotyped animals can be accurately imputed if they have genotyped porgenies. A combined reference population including individuals from othe populations can greatyly improves accuracy of imputation and genomic prediction due to increasing size of reference population
OriginalsprogEngelsk
UdgivelsesstedAU-Foulum
ForlagAarhus University, Faculty of Science and Technology
Antal sider85
ISBN (Trykt)978-87-92936-71-4
Rekvirerende organGraduate School of Science and Technology
StatusUdgivet - 3 jun. 2013

Note vedr. afhandling

Double degree PhD thesis: Beijing Agricultural University and Aarhus University

Se relationer på Aarhus Universitet Citationsformater

ID: 54243920