 |
 |
This is a repeat, but first time round I forgot to include some crucial information about money! Applications are invited for a BBSRC CASE PhD studentship starting in October 1999. Industrial CASE studentships are worth approximately 3,000 pounds more than a standard research council studentship. The student will be based in the Department of Statistical Science, University College London. The sponsoring company in PIC, the world's leading pig breeding organisation. The research topic is the analysis of high density genotyping information; in particular how to use the information from thousands of genetic markers to guide a breeding programme. Some more details are given below. For more information contact Dr Tom Fearn at UCL (email: [log in to unmask], tel: 0171 380 7189). To apply: send a CV to Dr Tom Fearn, Department of Statistical Science, University College London, Gower Street, London, WC1E 6BT. Applications will close on Tuesday 4th May. Topic: The analysis of High Density Genotyping information. In animal breeding, DNA information comes from a small number of genetic marker tests. Regression analysis is used to study the link between markers and traits of interest. With new DNA technology like GeneChip technology, the information per animal will come from 1000's of markers. This will require new approaches for data analysis. In general we deal with a trait of interest and a lot of information that is potentially related to that trait. The objective is to find linear and non-linear links between the trait and the mass of information. Other examples are fertility in males and a lot of information collected on the ejaculates or growth rate of individual animals and a lot of information collected about the feed, climatic conditions and health status. The objective of the project is to develop general solutions and to evaluate these with relevant cases as presented by the sponsor (PIC). The use of high density genotype information in animal breeding is the topic that will give direction to the project. The approaches explored are likely to include ideas from Bayesian statistics, from chemometrics, and possibly from artificial intelligence. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%