Linkage disequilibrium levels in Bos indicus and Bos taurus cattle using medium and high density SNP chip data and different minor allele frequency distributions

Linkage disequilibrium (LD), the observed correlation between alleles at different loci in the genome, is a determinant parameter in many applications of molecular genetics. With the wider use of genomic technologies in animal breeding and animal genetics, it is worthwhile revising and improving the current knowledge and understanding of cattle LD.This study analyzes levels of ID assessed through the r(2) measurement in seven breeds of cattle from both indicine (Bos indicus) and taurine (Bos taurus) (sub)species, genotyped with a high density panel (HD) of over 777,000 single nucleotide polymorphism (SNP) markers. Average levels of LD were assessed for inter-marker distances from 0 kb up to 10 Mb. Two more SNP panels were built in silico from the original genotypes to represent a lower density SNP chip including approximately 50,000 SNPs (50K) and a panel of 249,000 SNPs chosen to comply with specific minor allele frequency (MAF) distributions.The results show higher levels of ID than previously reported, especially at short inter-marker distances of less than 20 kb, assessed through the use of the higher density panel. Negligible impact of MAF changes in the estimation of r(2) was observed, supporting the use of the HD panel for LD-based methodologies even when ascertainment bias in the choice of SNPs is suspected. The 50K SNP panel indicated lower levels of LD in the first 100 kb and highly unreliable LD estimates for distances shorter than 50 kb. Overall, lower levels of LD were observed for all indicine breeds as compared to taurine breeds, with larger differences assessed in the first 100 kb, suggesting the need for use of higher SNP densities in this (sub) species when applying LD-dependent methods, as compared to most taurine breeds. (C) 2014 Elsevier B.V. All rights reserved.