During my PhD, it occurred to me that many of the papers in the scientific literature could be misidentifying genetic markers that show a skewed inheritance pattern, otherwise known as segregation distortion. This is because they used a statistical test that does not account for multiple testing, which increases the chance of false-positive results. Segregation distortion can be problematic for crop breeders, who wish to generate varieties with novel genetic compositions that are better suited to meeting agronomic objectives, such as increased yields or improved resistance to biotic or abiotic stresses. Distorted segregation in the genome could skew most lines in a population of crops away from the desired genotype, requiring breeders to create larger numbers of lines to compensate. It would be useful if we could attribute regions of segregation distortion to causative loci in the genome, as this would allow breeders to plan for the occurrence ahead of time.
I used simulations of meiosis, the process that generates egg and sperm cells, as well as empirical data from a population of wheat plants, to identify the ideal statistical test for detecting segregation distortion. The results show that many of the markers supposedly exhibiting segregation distortion in the literature are actually simply distorted by chance, rather than a genuine selection pressure operating during development. I also examined the effect of segregation distortion on the genetic mapping process, showing that only extreme distortion effects genetic mapping. These findings should inform future studies on best practices in both the identification and the use of genetic markers exhibiting segregation distortion, unlocking information on regions of the genome that were previously hidden to researchers.
Alex Coulton, SWBio DTP Student
Paper: Segregation distortion: Utilizing simulated genotyping data to evaluate statistical methods by Alexander Coulton, Alexandra M. Przewieslik-Allen, Amanda J. Burridge, Daniel S. Shaw, Keith J. Edwards, Gary L. A. Barker in PLoS ONE