Dissecting the Genetics and Phenotypic Effects which Diversify Canine Skeletal Morphology

Supervisors: Jeffrey J. Schoenebeck, Dylan Clements

Project description:


The morphological diversity of the dog skeleton is unparalleled by other animal species. Increasing evidence suggests that mobile repeat elements and noncoding DNA are drivers of both evolution and disease. Recently we described a transposon that is associated with canine brachycephaly, the breed-defining shortening of the face that is common to dog breeds like the English bulldog and Pug1,2. The transposon inserted between the exons of SMOC2, a gene whose protein product remains poorly understood. Given SMOC2’s expression elsewhere in the developing skeleton, we predict that its disruption effects the postcranial skeleton as well.


This PhD project will provide our trainee with the intellectual support and the resources to develop image analyses and bioinformatics skills, and their integration with patient data. Data generation will span both dry and wet labs, thus providing the trainee with a broad exposure to laboratory techniques. Project goals include:

  1. Develop semi-automated segmentation of computed tomography reconstructions. Bioimaging is a rich source of “Big Data” that requires new approaches to make it applicable to population studies. The trainee will assess new methodologies developed by the Schoenebeck group to segment and analyse CT image reconstructions of the canine skeleton, such as the pelvis.
  2. Test whether transposon insertion into SMOC2 affects pelvis form. Morbidities pertaining to the pelvis reduce brachycephalic dogs’ healthspan. Pelvis phenotypes (Objective 1) will be tested for association with markers of skeleton scale and shape. The SMOC2 transposon insertion will be included in this analysis, to determine whether it is associated with the distinctive features of pelvis morphology that are observed among brachycephalic dogs.
  3. Identify putatively causal variants that are associated with bone phenotypes. A major challenge of canine genetics research is the suboptimal quality of its draft genome assembly, which hinders identification of key genetic variants that underlie skeletal phenotypes and disease risk. To circumvent shortcomings of the dog assembly, the trainee will develop and deploy targeted long read resequencing to generate regionalised de novo assemblies from which short reads of strategically resequenced dog genomes will be aligned for variant calling. This approach will be applied to a number of outstanding quantitative trait loci including the IGF1 locus, the latter whose phenotypic effects are associated with reducing the scale and shape of canine face length.
  4. Develop models of genetic variant contributions to skeletal morphology for clinical application. Ultimately the clinical application of molecular phenotyping has potentially important clinical applications in dog breeds, for example for identifying individuals whom might be at greater risk of dystocia because of their pelvic morphology. Using the large patient record database, imaging and DNA archives, the contributions of variants to pelvic morphology will be tested to develop genotype-based risks models.


1.       Marchant, T. W. et al. Canine Brachycephaly Is Associated with a Retrotransposon-Mediated Missplicing of SMOC2. Curr. Biol. 27, 1573–1584.e6 (2017).

2.       Schoenebeck, J. J. et al. Variation of BMP3 Contributes to Dog Breed Skull Diversity. PLoS Genet 8, e1002849 (2012).