Developing flexible models for genetic evaluations in smallholder crossbred dairy farms

Area of Science:

• Animal Genetics
• Quantitative Genetics
• Dairy Science

Background:

• Smallholder dairy farms in developing countries exhibit low productivity.
• Genomic selection offers potential for genetic gains but requires tailored evaluations for limited pedigree and small farm sizes.

Purpose of the Study:

• To propose and validate a flexible Bayesian model (BayesR) for genetic evaluation of milk yield in smallholder dairy farms.
• To assess the model's ability to handle nongenetic random effects and varying SNP variance.

Main Methods:

• Simulations using Indian crossbred dairy cattle genotype data.
• Application of the BayesR model to real data from 4,655 crossbred cows with 106,109 milk records and 689,750 SNPs.
• Cross-validation for accuracy and bias estimation of genomic estimated breeding values (GEBV).

Main Results:

• The BayesR model effectively separated genetic and nongenetic parameters in small herds.
• GEBV accuracy increased with farm size up to approximately 5 cows.
• Estimated heritability for milk yield was 0.16, with GEBV accuracy of 0.45 and bias of 1.04.
• Candidate genes IMMP2L and ARHGEF2 were identified near top genetic variants.

Conclusions:

• The proposed Bayesian model is suitable for genetic evaluation in smallholder dairy systems.
• Low heritability and GEBV accuracy highlight the need for larger phenotyped and genotyped populations.
• Increasing herd size (3-5 cows) can improve genetic effect estimation in these systems.