Progress 08/01/13 to 03/20/15
Outputs
Target Audience: Academic veterinarians Practicing veterinarians Horse owners Veterinary students Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?
Nothing Reported
How have the results been disseminated to communities of interest? Scientific meetings with abstracts 1. Finno CJ, Valberg SJ, Peterson J, Draper ACE, , Livesey L, Baird J, Firshman AM. Genome-wide association analysis of Shivers in Belgian horses. Plant and Animal Genetics Conference. San Diego CA 2013 2. Finno CJ, Valberg SJ, Mickelson JM. Whole genome next-generation sequencing of 4 Belgian horses. College of Veterinary Medicine Research Day October 2013 Veterinary meetings Valberg SJ Movement Disorders in Horses: Shivers, Stringhalt and other conundrums. California Bay Area Equine Practitioners. October 2012 Valberg SJ. Shivers in Horses. Purina Equine Conference. October 2013 Valberg SJ. Shivers, Stringhalt and other Movement Disorders SIVE, Mila Italy February 2014 Hores owner magazines Spring 2014. University of Minnesota Medical Bulletin. From humans to horses January 2015: Equus magazine. Feature article on our research on Shivers research What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Next-generation sequencing identified 13.3 million variants (12 million SNPs and 1.3 million indels). Of these variants, 65% were intergenic, 30% were intronic, 2.5% were exonic and 2.5% were 500bp upstream or downstream of an annotated gene. The region on ECA8 that contained the associated SNPs from the GWAS (59.34-59.72 Mb; 382 kb in length) was evaluated first to determine if any of the variants segregate with the disease phenotype. Within this region, there were no variants that segregated with the disease phenotype, thereby excluding ECA8:59.34-59.72 Mb as a putative region for a genetic mutation for Shivers. To evaluate additional genomic regions for an association with the Shivers phenotype, an association analysis was performed across all 13.3 million detected variants. Using the program SNPSift Case control, each horse was labeled according to its phenotype and a Fisher Exact Test using four different models (dominant, recessive, allelic and genotypic) was performed for each of the 13.3 million variants. The allelic model was evaluated for significant associations of coding variants. Using a cut-off significance level of P=0.05 and selecting only non-synonymous variants, 88 variants were found to segregate with the phenotype whereby the affected horses possessed the non-reference allele. Haplotypes were identified within these 88 variants and 57 regions were prioritized for further assessment. Genotyping was performed on these variants using Sequenom genotyping technology on 43 affected Belgian horses and 66 controls. A non-synonymous coding mutation was identified in PMS1, a gene associated with DNA repair. All affected cases were homozygous for the missense mutation and 3/5 control horses were heterozygous for this mutation. The mutation resulted in a conservative valine to leucine amino acid change. Further investigation of this mutation is currently underway in additional Shivers-affected horses. Additionally, a custom pipeline has been designed, using FreeBayes as a variant-calling algorithm, to effectively identify larger (>10bp) indels in this next-generation sequence data and this algorithm will be run on this dataset.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2014
Citation:
1. Draper ACE, Bender JB, Firshman AM, Baird JD, Reed S, Mayhew IJ and Valberg SJ. Epidemiology of characterisation
of Shivers in horses. Equine Vet J 2014 May 6. doi: 10.1111/evj.12296. [Epub ahead of print]
2.Draper ACE, Trumble, TN, Firshman AM, Baird JD, Reed S, Mayhew IJ and Valberg SJ Posture and movement
characteristics of forward and backward walking in horses with Shivers and acquired bilateral Stringhalt. Equine Vet J. 2014
Mar 10. doi: 10.1111/evj.12259. [Epub ahead of print]
|
Progress 08/01/13 to 09/30/13
Outputs
Target Audience:
Nothing Reported
Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided? A DVM student worked on this project throughout the year as part of her postdoctoral research experience. How have the results been disseminated to communities of interest? Data was reported at the University of Minnesota Research Day and genome workshops such as Havemeyer 2013 and PAG 2013. What do you plan to do during the next reporting period to accomplish the goals? The next step is a second Sequenom analysis to examine the remaining 41 regions, including SNVs on ECAX and ECA18 in additional breeds. These regions may still be associated with Shivers but have not achieved significance due to a high allele frequency in the Belgian draft horse breed.
Impacts
What was accomplished under these goals?
Our original plan was to select 3 Shivers-affected Belgian geldings and 3 age-matched control Belgian geldings; however, we combined funding from USEF and UMEC for Shivers to expand the number of horses to 5 Shivers-affected Belgian geldings and 5 age-matched control Belgian geldings and progress is reported for the combined analysis. DNA from ten horses was submitted to the Biomedical Genomic Center at the University of Minnesota, where the DNA was fragmented to create a DNA library for each horse and whole-genome next generation sequencing was performed on 100 bp paired-end reads using the Illumina HiSeq2000 sequencing machine. After performing quality control on the sequences obtained, we had, on average, 184 million reads/horse (13.6x coverage of genome). Of these reads, 98% of reads mapped successfully to the equine reference genome, with 96% properly paired. The genome analysis tool kit (GATK) software package was used to call variants (single nucleotide variants or SNVs and insertions/deletions or indels). 16 million variants were identified using a quality cut-off score of 4. Fifteen million SNVs and 1 million indels were identified. Of these, 65% were intergenic, 30% were intronic, 2.5% were exonic, 2.5% were 500 bp upstream ordownstream of an annotated gene. A case/control Fisher Exact Test was performed on variants; using an allelic model with a cut-off of <0.001, 133 variants were identified and using an allelic model with a cut-off of <0.01 20,179 variants were identified. Based on these results, we excluded the original region on ECA8 (59.3-57.72). We had an excellent average quality score (phred score) for variant calls of 815 (phred score of 50 has a call accuracy of 99.999%) and 57 regions were identified in which variants were associated with a haplotype block. A Sequenom assay (Illumina, San Diego, CA) using 43 Belgian draft horses with Shivers and 66 controls was performed for 63 of the variants identified by next generation whole genome sequencing. Four variants failed the multiplex design, 10 were not informative (not polymorphic) and 49 variants remained (13 non-synonymous, 13 intronic, 23 intergenic). An association analysis and logistic regression with sex as a covariate were performed. One SNV on the X chromosome (chrX_70406663; unadjusted p value of 0.02 for regression) was identified but no gene or transcript was found in this region. Two SNVs on ECA18 (66320637, 66321297; unadjusted p value of 0.06 for regression) were also identified. These SNVs are located within a gene that is involved in DNA repair. One of the SNVs causes a non-synonymous substitution within this gene. Although these values approached significance, based on the number of multiple test performed, an adjusted p value of 0.001 would be considered significant for this analysis. This first Sequenom analysis excluded 16 regions with variants.
Publications
|
|