A genome-wide scan for signatures of differential artificial selection in ten cattle breeds

Background: Since the times of domestication, cattle have been
continually shaped by the influence of humans. Relatively recent
history, including breed formation and the still enduring enormous
improvement of economically important traits, is expected to have
left distinctive footprints of selection within the genome. The
purpose of this study was to map genome-wide selection signatures
in ten cattle breeds and thus improve the understanding of the
genome response to strong artificial selection and support the
identification of the underlying genetic variants of favoured
phenotypes. We analysed 47,651 single nucleotide polymorphisms
(SNP) using Cross Population Extended Haplotype Homozygosity
(XP-EHH). Results: We set the significance thresholds using the
maximum XP-EHH values of two essentially artificially unselected
breeds and found up to 229 selection signatures per breed. Through
a confirmation process we verified selection for three distinct
phenotypes typical for one breed (polledness in Galloway, double
muscling in Blanc-Bleu Belge and red coat colour in Red Holstein
cattle). Moreover, we detected six genes strongly associated with
known QTL for beef or dairy traits (TG, ABCG2, DGAT1, GH1, GHR and
the Casein Cluster) within selection signatures of at least one
breed. A literature search for genes lying in outstanding
signatures revealed further promising candidate genes. However, in
concordance with previous genome-wide studies, we also detected a
substantial number of signatures without any yet known gene
content. Conclusions: These results show the power of XP-EHH
analyses in cattle to discover promising candidate genes and raise
the hope of identifying phenotypically important variants in the
near future. The finding of plausible functional candidates in some
short signatures supports this hope. For instance, MAP2K6 is the
only annotated gene of two signatures detected in Galloway and
Gelbvieh cattle and is already known to be associated with carcass
weight, back fat thickness and marbling score in Korean beef
cattle. Based on the confirmation process and literature search we
deduce that XP-EHH is able to uncover numerous artificial selection
targets in subpopulations of domesticated animals.