# Genomic Prediction of Milk Fat Percentage Among Crossbred Cattle in the Indian Subcontinent

## TL;DR

This study uses genomic data to predict and improve milk fat percentage in crossbred dairy cows in India, offering a tool for breeders to select high-fat-producing animals.

## Contribution

The study introduces a genomic prediction method tailored for crossbred cattle in smallholder systems in the Indian subcontinent.

## Key findings

- Genomic data imputation improved allele frequency correlation from 0.594 to 0.882.
- Heritability of milk fat percentage was estimated at 0.10, suggesting limited genetic potential for improvement.
- GEBV ranged from 0.096 to 3.10%, providing a practical tool for selecting high-fat-producing cows.

## Abstract

This study was conducted to improve the milk fat percentage for crossbred dairy cows in smallholder systems, using genomic-estimated breeding value (GEBV). The phenotypic data were collected across six states in India from the crossbred cows with varied level of exotic inheritance. The genetic analysis was carried out using 50k SNP BeadChip, and imputation improved the accuracy of genomic data, boosting allele frequency correlation. The analysis was conducted using FImpute software version 2.2. Heritability was estimated using Bayes R, suggesting cautious use for breeding improvements. The GEBV offers breeders a practical tool for selecting high-fat-yielding cows. This research provides insights into enhancing milk fat percentage and genetic improvement strategies in smallholder dairy systems.

This study focused on improving the milk fat percentage for crossbred dairy cows in smallholder systems, using GEBV. The data were collected from 2507 animals between 2016 and 2023 under BAIF’s Enhanced Genetic Gains program in Pune, India. After refining the dataset, 33,845 records from 1896 animals were analyzed. The result showed that 75.54% of farms had either one or two animals. Prior to quality control, the mean milk fat percentage was 3.94%, but it decreased to 3.83% after data pruning, which necessitated removing the outliers. Genetic analysis involved 1478 animals genotyped for 49,911 SNPs after applying a rigorous quality control process, and imputation improved the accuracy of genomic data, boosting allele frequency correlation from 0.594 to 0.882. The study revealed that the additive genetic variance, phenotypic variance, and error variance were calculated as 0.012, 0.118, and 0.106, respectively. The heritability was estimated at 0.10, suggesting cautious use for breeding improvements. The GEBV ranged from 0.096 to 3.10%, which offers breeders a practical tool for selecting high-fat-producing cows. This research provides valuable insights into optimizing milk quality and advancing genetic improvement strategies in smallholder dairy systems.