Daniel Tolhurst


Contact details



The Roslin Institute
Easter Bush Campus

Post code
EH25 9RG


My PhD project will focus on efficient estimation within the linear mixed model (LMM) paradigm. These models have numerous applications to many fields in the biological sciences, particularly animal, plant and human genetics. Motivation for the project has stemmed from the need to analyse complex genomic data in real time within global plant breeding programmes. This requires the efficient estimation of genetic variance parameters and subsequent prediction of genetic effects (value) for multi-environment trial (MET) datasets. These datasets typically represent a cumulation of field data from a larger number of locations and years. Estimation becomes very computationally demanding for problems involving large-scale genomic MET data coupled with an extensive number of random effects, which is becoming more prevalent in global breeding programmes and plant breeding in general. As such, there is high demand for efficient and accurate estimation under the LMM compared to traditional computer algorithms. In this project, we will implement efficient methods of estimation based on Markov Chain Monte Carlo (MCMC), Gibbs Sampling and other iterative schemes to handle dense genomic MET data and complex variance structures at both the residual and genetic level.


2009-2014: BSc (Hons) Medical Mathematics

Thesis: Imputation as a Method of Estimation in Linear Mixed Models