Plant & Animal Genomes XVII Conference

January 10-14, 2009
Town & Country Convention Center
San Diego, CA

Genetic And Physiological Simulation Of Crossing And Selection Strategies To Combine Diagnostic Markers And Quantitative Traits

Scott Chapman¹ , Jiankang Wang² , Karine Chenu³ , Greg Rebetzke⁴ , David Bonnett⁵ , Claude Welcker⁶ , Francois Tardieu⁷ , Mark Dieters⁸ , Graeme Hammer⁹

¹  CSIRO Plant Industry, 306 Carmody Rd, St. Lucia, QLD, 4065, Australia

²  CIMMYT at Institute of Crop Science CAAS, Beijing, 100081, China

³  APSRU Department of Primary Industries and Fisheries, 203 Tor St, Toowoomba, QLD, 4350, Australia

⁴  CSIRO Plant Industry, P.O. Box 1600, Canberra, ACT 2601, Australia

⁵  CIMMYT, Apdo. Postal 6-641, Mexico DF, 06600, Mexico

⁶  INRA, UMR 759 LEPSE, 2 place Viala, 34060 Montpellier cedex 01, France

⁷  The University of Queensland, Brisbane, Qld 4072, Australia

Diagnostic molecular markers are available for more than 20 traits in wheat. The challenge is to combine these markers with phenotypic selection in creating new parental lines and progeny (target genotypes). Based on previous simulation work, we considered six diagnostic markers related to height, grain quality and disease. The objective was to integrate diagnostic selection (including Rht-8, a major gene to increase coleoptile length) with selection for an additional six minor QTL for this quantitative trait. The approach is to simulate (http://www.uq.edu.au/lcafs/qugene/) crosses between wheat lines with known genotypes for the markers so that the effects of linkage (among genes and between genes and markers) can be accounted for during the process of selection. We aim to identify the most efficient crossing and marker-aided process to quickly combine the desired genes together. In this case, production of doubled haploids following F2 enhancement for diagnostic markers was most efficient. In an extension of this research, we are integrating physiological models into the simulation. For each genotype to be simulated, the gene effects are linked to ‘physiological traits’ (e.g. leaf growth response to temperature or water deficit) that are the inputs to a deterministic physiological model. This approach allows the ‘scaling up’ of organ-level understanding of genetic response to a crop-level response. For a given QTL mapping study of leaf growth, we were able to estimate the potential outcomes of these QTL effects on the yield of maize in a range of different temperature and water deficit environments.