The International Complex Trait Consortium

High Resolution Mapping of QTL Using a Coalescent Approach

Jonathan Flint1, Anjela Meesaq1, Christopher Talbot1, Jan Fullerton1, Robert Deacon2, Nicholas Rawlins2, Andrew Morris1

1Wellcome Trust Centre for Human Genetics, Roosevelt Drive, Oxford OX3 7BN
2Department of Experimental Psychology, University of Oxford, South Parks Road, Oxford

In order to develop methods that map QTL to the point where positional cloning is feasible, we have been developing methods that use outbred animals. The potential of standard laboratory outbred mice has not been explored for QTL mapping, even though it is likely that the extent of linkage disequilibrium in these mice would make high resolution mapping by association a realistic option. We have been working on a mouse model of anxiety using quantitative, correlated measures (including activity and defecation in an open-field arena (abbreviated to OFA and OFD)). We have already mapped QTL influencing these phenotypes into regions of less than 1 cM on chromosomes 1 and 15. In order to increase the resolution further we have been using MF1 mice (acquired from Harlan, UK, Ltd). We show that it is possible to detect QTL by association in these mice and we have found that linkage disequilibrium extends for up to 100 Kb. However single locus association tests cannot be efficiently utilised to refine location within regions of positive genetic linkage. This has prompted the development of methodologies that combine information across linked markers. We have used a novel coalescent method developed by Andrew Morris at the Wellcome Trust Centre to deal with this problem. Using three generation pedigrees of MF1 mice we show how the shattered coalescent method can produce 95% confidence intervals of about 500 Kb, and enables us to identify a small number of candidate genes for functional studies.