Genomic prediction of crossbred dairy cattle in Tanzania: A route to productivity gains in smallholder dairy systems

摘要

Selection based on genomic predictions has becomethe method of choice for genetic improvement in dairycattle. This offers huge opportunity for developingcountries with little or no pedigree data, and preliminarystudies have shown promising results. The AfricanDairy Genetic Gains (ADGG) project initiated a digitalsystem of dairy performance data collection, accompaniedby genotyping in Tanzania in 2016. Currently,ADGG has the largest body of dairy performance datagenerated in East Africa from a smallholder dairy system.This study examines the use of genomic best linearunbiased prediction (GBLUP) and single-step (ss)GBLUP for the estimation of genetic parameters andaccuracy of genomic prediction for daily milk yield andbody weight in Tanzania. The estimates of heritabilityfor daily milk yield from GBLUP and ssGBLUP wereessentially the same, at 0.12 ± 0.03. The heritabilityestimates for daily milk yield averaged over the wholelactation from random regression model (RRM) GBLUPor ssGBLUP were 0.22 and 0.24, respectively.The heritability of body weight from GBLUP was 0.24± 04 but was 0.22 ± 04 from the ssGBLUP analysis.Accuracy of genomic prediction for milk yield from aforward validation was 0.57 for GBLUP based on fixedregression model or 0.55 from an RRM. Correspondingestimates from ssGBLUP were 0.59 and 0.53, respectively.Accuracy for body weight, however, was muchhigher at 0.83 from GBLUP and 0.77 for ssGBLUP.The moderate to high levels of accuracy of genomicprediction (0.53–0.83) obtained for milk yield and bodyweight indicate that selection on the basis of genomicprediction is feasible in smallholder dairy systems andmost probably the only initial possible pathway toimplementing sustained genetic improvement programsin such systems.