copy number variation detection in sheep genome by using ovine beadchip 50k

introduction recently, genomic research in livestock is focused on genomic variation and its effect on phenotypic performance in economic traits. copy number variation (cnv) is one of these variations in genome including insertion, deletion and duplication of 1 kb to 1 mb segment with more than 90% similarity. cnvs can change gene structure and dosage, can regulate gene expression and function and (1, 4). in mammals, it is important source of variability in genomes and it contains 0.4-25% of whole genome variation. some researches carried out in livestock have been demonstrated that cnv affecting genes or gene regions are associated with several phenotypic traits. for example, cnv in intron 1 of the sox5 gene causes the pea-comb phenotype in chicken. cnv affects also the agouti locus in sheep and goats and contributes to the variability of coat color in these two species. the late feathering locus in this avian species includes a partial duplication of the prlr and spef2 genes and dominant white locus in pigs includes alleles determined by duplications of the kit gene (2, 5, 6). in spite of, many researches carried out in human represent association between cnv with both complex genetic diseases and traits; however, far too little attention has been paid to cnv in farm animal. this paper will focused on detecting of cnv in sheep genome. materials and method the sheep genomic dna was extracted from blood of 360 italian ewes using dna purification kit (promega corporation, madison, wi). markers were genotyped by illumina ovinesnp50 beadchip according to instructions. it is containing 54,241 markers that uniformly span the entire ovine genome (illumina, inc., usa). after completion of the assay, the beadchips were scanned with a two-color, confocal bead array reader. scanned image intensities were loaded directly into illumina’s beadstudio 1.2 software. when normalization was completed, the clustering process was performed to assess cluster position for each marker and to determine individual genotypes. lrr and baf of sample were reported. the pfb file was calculated based on the baf of each marker in these populations. the sheep gc model file was generated by calculating the gc content of the1 mb genomic region surrounding each marker (500 kb each side). cnvs were inferred using a penncnv (http: //www.openbioinformatics. org/penncnv/). penn cnv quality filters were applied after cnv detection. high quality samples with a standard deviation (sd) of lrr < 0.30 and with the default set: baf drift as 0.01 and waviness factor value between − 0.05 and 0.05, were used respectively. in addition, the program argument: the “lastchr 26” in the “detect” argument were used for specific cnvs. cnvrs were determined by aggregating overlapping cnvs identified in different animals. the ucsc table browser tool was used to identify the gene content located within or partially overlapping with the cnvrs and david bioinformatics resources (http://david.abcc.ncif crf.gov) was used for further go functional analysis, including gene ontology. results and discussion after all filtration 184 samples were remained. all cnvs and cnvrs found in one sample were omitted from further analysis. finally 904 cnvs (599 losses, 111 gains and 194 losses/gains) were detected. the average number of cnvs per sample was 4.91, with an average length and median size of 170 kb and 123.9 kb, respectively. 60% of all cnvs had length between 100 kb to 500 kb. this result was similar to other research (2, and 3). after all cnvs were aggregated for the cnv region (cnvr), 88 cnvrs were identified that 55 event were found just in one sample and were omitted. the average and median size of cnvr were 178.61 kb and 135.25 kb. the profile of cnvrs location on ovis_aris_3.0 genome were shown distribution of cnvrs was not randomly. the highest percentages covering of cnvrs located on chromosomes 16, 24, 25 and 26 (0.8%, 0.9%, 0.8% and 2%, respectively). it’s similar to result of liu et al (21). gene ontology (go) analysis can provide insight into the functional enrichment of cnvs. for this reason, we ran go analysis using david http://david.abcc.ncifcrf.gov. two cnvrs (chr12:481088-180295, chr16: 385305-156698) entirely encompassing myog refseq gene and mi103 respectively. the gene content of the 25 cnvr s, we used a blastn search for homologous human and cattle sequences using the ucsc table browser tool. there were 110 refseq homologous human genes located within or partially overlapping with 16 cnvrs and similarly, there were 40 refseq homologous cattle genes located within or partially overlapping with 10 cnvrs. conclusion comparing cnvs and cnvrs identified in sheep genome with cnvrs reported in cattle showed demonstrated low level of similarity, so this genomic variation had great potential detection and using in breeding scheme in sheep industry.