Preprint compares SNP selection strategies for reduced-density genomic panels in beef cattle

A preprint posted to bioRxiv describes a systematic comparison of five SNP selection strategies for constructing reduced-density genotyping panels suited to routine genomic evaluations in beef cattle.

As the number of genotyped animals in livestock breeding programmes has grown into the millions, building the genomic relationship matrices used in genomic selection has become computationally demanding. High-density SNP chips — while capable in principle of capturing causal variants — impose substantial computational costs without proportional gains in prediction accuracy for most traits. The authors compared strategies including random selection, equal-interval spacing across chromosomes, MAF-based filtering, linkage-disequilibrium pruning, and a method prioritising variants with the largest estimated effects from prior analyses.

Across the traits examined, the study found that informed SNP selection approaches — particularly those prioritising variants with known or estimated effect sizes — maintained genomic prediction accuracy comparable to full high-density panels while using markedly fewer markers. The work contributes to an active area of livestock genomics research aimed at making large-scale genomic evaluation computationally tractable as herd databases expand.

This work has not yet been peer-reviewed. The preprint is available at bioRxiv (DOI: 10.64898/2026.08.26.747408). It is primarily relevant to researchers and educators working in quantitative genetics, livestock genomics, and animal breeding.