Genome-wide detection of copy number variations among diverse horse breeds by array CGH

Abstract

Recent studies have found that copy number variations (CNVs) are widespread in human and animal genomes. CNVs are a significant source of genetic variation, and have been shown to be associated with phenotypic diversity. However, the effect of CNVs on genetic variation in horses is not well understood. In the present study, CNVs in 6 different breeds of mare horses, Mongolia horse, Abaga horse, Hequ horse and Kazakh horse (all plateau breeds) and Debao pony and Thoroughbred, were determined using aCGH. In total, seven hundred CNVs were identified ranging in size from 6.1 Kb to 0.57 Mb across all autosomes, with an average size of 43.08 Kb and a median size of 15.11 Kb. By merging overlapping CNVs, we found a total of three hundred and fifty-three CNV regions (CNVRs). The length of the CNVRs ranged from 6.1 Kb to 1.45 Mb with average and median sizes of 38.49 Kb and 13.1 Kb. Collectively, 13.59 Mb of copy number variation was identified among the horses investigated and accounted for approximately 0.61% of the horse genome sequence. Five hundred and eighteen annotated genes were affected by CNVs, which corresponded to about 2.26% of all horse genes. Through the gene ontology (GO), genetic pathway analysis and comparison of CNV genes among different breeds, we found evidence that CNVs involving 7 genes may be related to the adaptation to severe environment of these plateau horses. This study is the first report of copy number variations in Chinese horses, which indicates that CNVs are ubiquitous in the horse genome and influence many biological processes of the horse. These results will be helpful not only in mapping the horse whole-genome CNVs, but also to further research for the adaption to the high altitude severe environment for plateau horses.

Figure 1. Distribution of CNVs in different breeds.

(A) Distribution of breed-specific and shared CNVs among diverse horse breeds. The red represents breed-specific CNVs. The blue represents shared CNVs, and the color from light to dark indicates CNVs shared between 2, 3, 4 and 5 samples. (B) Percentage of CNVs shared among samples. The red represents 1 sample. The blue represents 2 samples. The green represents 3 samples. The purple represents 4 samples. The dark blue represents 5 samples.

Figure 2. Distribution of CNVRs in the horse genome.

Numbers in parentheses represent CNVRs. The red represents gain CNVRs. The light blue represents loss CNVRs. The yellow represents both CNVRs.

Figure 3. Chromosome distribution of the number of CNVRs.

Figure 4. Correlation between the number of CNVRs and chromosome length.

Figure 5. Size range distribution of the CNVRs detected.

Figure 6. Percentage of gene biotypes affected by CNVs.