Open Access Open Badges Research article

Molecular evolution of the Bovini tribe (Bovidae, Bovinae): Is there evidence of rapid evolution or reduced selective constraint in Domestic cattle?

Sean MacEachern125*, John McEwan3, Alan McCulloch3, Andrew Mather1, Keith Savin1 and Mike Goddard14

Author Affiliations

1 Primary Industries Research Victoria, Animal Genetics and Genomics, Attwood, VIC 3049, Australia

2 Latrobe University, Department of Genetics, Bundoora, VIC 3086, Australia

3 Animal Genomics, AgResearch, Invermay, Private Bag 50034, Mosgiel, New Zealand

4 Melbourne University, School of Agriculture and Food Systems, Melbourne VIC 3000, Australia

5 Avian Disease and Oncology Laboratory 3606 E Mt Hope Rd, East Lansing, MI 48823, USA

For all author emails, please log on.

BMC Genomics 2009, 10:179  doi:10.1186/1471-2164-10-179

Published: 24 April 2009



If mutation within the coding region of the genome is largely not adaptive, the ratio of nonsynonymous (dN) to synonymous substitutions (dS) per site (dN/dS) should be approximately equal among closely related species. Furthermore, dN/dS in divergence between species should be equivalent to dN/dS in polymorphisms. This hypothesis is of particular interest in closely related members of the Bovini tribe, because domestication has promoted rapid phenotypic divergence through strong artificial selection of some species while others remain undomesticated. We examined a number of genes that may be involved in milk production in Domestic cattle and a number of their wild relatives for evidence that domestication had affected molecular evolution. Elevated rates of dN/dS were further queried to determine if they were the result of positive selection, low effective population size (Ne) or reduced selective constraint.


We have found that the domestication process has contributed to higher dN/dS ratios in cattle, especially in the lineages leading to the Domestic cow (Bos taurus) and Mithan (Bos frontalis) and within some breeds of Domestic cow. However, the high rates of dN/dS polymorphism within B. taurus when compared to species divergence suggest that positive selection has not elevated evolutionary rates in these genes. Likewise, the low rate of dN/dS in Bison, which has undergone a recent population bottleneck, indicates a reduction in population size alone is not responsible for these observations.


The effect of selection depends on effective population size and the selection coefficient (Nes). Typically under domestication both selection pressure for traits important in fitness in the wild and Ne are reduced. Therefore, reduced selective constraint could be responsible for the observed elevated evolutionary ratios in domesticated species, especially in B. taurus and B. frontalis, which have the highest dN/dS in the Bovini. This may have important implications for tests of selection such as the McDonald-Kreitman test. Surprisingly we have also detected a significant difference in the supposed neutral substitution rate between synonymous and noncoding sites in the Bovine genome, with a 30% higher rate of substitution at synonymous sites. This is due, at least in part, to an excess of the highly mutable CpG dinucleotides at synonymous sites, which will have implications for time of divergence estimates from molecular data.