RESEARCH >> Experimental Design
Most biological traits that affect profitability for farms, forests or horticultural crops are polygenic in nature. That is, they are influenced by many genes, each having a small effect. In order to detect these small effects we need experimental designs that have sufficient power, are logistically feasible and cost-effective.
The aim is to develop experimental designs to enable effective gene-discovery in QTL and association studies across the range of gene mapping applications in New Zealand. We will develop methods to construct and evaluate experimental designs capable of effectively detecting genomic associations. The designs constructed will have good power (e.g. 0.9) to detect effects of a given size with a sufficiently high Bayes factor (i.e. the ratio of how much more likely the data are if we assume there is a real effect, than if there is none), to overcome the low prior odds for any given SNP to be, or to be closely associated with, a functional locus.
Experimental design end-users will have assurance that the frameworks are robust since the optimal number of genotypes, clonal replicates, and QTL, association mapping sample sizes and experimental layouts were determined in realistic New Zealand-based scenarios.
KEY PERSONNEL
Statisticians
Rod Ball, Scion, Lead Statistician (Bayesian Statistician)
Ken Dodds, AgResearch, Lead Statistician
Benoit Auvray, AgResearch, Statistician
Mik Black, University of Otago, Lead Statistician and Bioinformaticist
Geneticists
Phillip Wilcox, Scion, Lead Geneticist, and Project Governance Group member and Project Leader
David Chagne, Plant and Food Ltd. Geneticist
Rod Lea, ESR Ltd. Geneticist
PhD Student
