Applications are now open for the following PhD studentship in Statistical Health Genetics at the Roslin Institute:
Background
Individuals differ widely in their response to infection. Whereas resilient individuals manage to recover quickly and fully, less resilient individuals may experience prolonged infection with long-term health damage, and some may even die. Resilience to infection is an important trait in all species, not only because it affects an individual’s own health, but also because an individual’s resilience is likely to affect its ability to transmit infection and thus impact on the health of others.
The recent explosion of biological data and innovative data-driven computational methods provide exciting new opportunities to monitor and analyse individuals’ infection trajectory over time, and determine to what extent these are genetically regulated. However, adequate mathematical and computational methods to quantify and analyse individuals’ resilience trajectories are currently lacking.
Aims:
The aims of this PhD studentship are to
(1)Apply mathematical Hidden Markov Models and computational Deep Learning methods to characterise resilience trajectories of individuals and identify different response types
(2) To determine to what extent resilience trajectories are genetically regulated and dissect their underlying genetic architecture
(3) To develop a genetic-epidemiological prediction model to assess the feasibility of integrating resilience trajectories into genetic selection, and it’s impact on individual and herd health.
Approach:
We will use a large and unique existing dataset comprising longitudinal disease data together with ‘omics’ data from thousands of genotyped pigs that were experimentally challenged with a virulent strain of the Porcine Reproductive and Respiratory Syndrome virus (PRRSV) to construct and analyse resilience trajectories. To analyse the resilience trajectories, we will build upon previously developed mathematical methods (Lough et al. Proc. R. Soc B. 2015; Torres et al. 2016. Plos Biol. 14(6)), and complement these with Hidden Markov Models and Deep Learning methods to characterise trajectories and trajectory phenotypes of resilient animals. We will then use random regression models combined with genome wide association studies to determine the genetic architecture of resilience trajectories. Finally, we will use genomic prediction models coupled with epidemiological models to assess the use of resilience trajectories to breed healthier animals, and the requirements for data recording and beneficial consequences of doing so.
Project team & Training:
Supervisors:
Prof. Andrea Doeschl-Wilson (The Roslin Institute, University of Edinburgh)
Prof. Ruth King (School of Mathematics, University of Edinburgh)
Dr Ricardo Pong-Wong (The Roslin Institute, University of Edinburgh)
Project partners:
Prof. Jack Dekkers (Iowa State University, USA) and Dr. William Herring (Pig Improvement Company, PIC)

The PhD student will join a vibrant multi-disciplinary team at Roslin consisting of bio-mathematicians and geneticists, and will receive training in mathematical and computational modelling, as well as animal breeding and genetics, both through targeted course and close individual supervision. The successful candidate will have the opportunity to help resolve one of the most pressing livestock disease problems by closely collaborating with world leading livestock geneticists from industry and academia.
This project builds upon and extents a long-standing international collaboration on animal health genetics between the Roslin Institute, Iowa State University (Prof. Jack Dekkers), and the industrial partner Genus. There is opportunity for short-term placements with the project partners.

Further information / How to apply

https://www.ed.ac.uk/roslin/postgraduate/studentships/genetic-dissection-of-individual-health-trajectories-and-their-role-in-disease-transmission

Application Deadline: January 20th 2018


The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.