P6.28 Poster Session — Monday 29th June 2009 Crop modelling as an aid for environmental characterisation and crop improvement Karine Chenu (DPIF), Scott C. Chapman (CSIRO), Greg McLean (DPIF), Douglas Lush (DPIF), Graeme L. Hammer (University of Queensland), Fernanda Dreccer (CSIRO) Genotype–environment interactions impede crop improvement for complex traits such as yield in water limited environments. Crop modelling is a powerful tool to characterise the nature of the environmental stresses experienced across locations, years, manage- ments and genotypes. Simulation of the seasonal plant environment can aid in understanding genotype–environment interactions within and across trials and in identifying key adaptive traits specific to different environment types. An environmental characterisation has been undertaken for wheat in North East Australia. Simulations based on more than 100 years climatic data were conducted for representa- tive sites, soils and management systems, for the cultivar Hartog. Five environment types based on patterns of crop water stress around flowering were identified and the frequency of occurrence of each environment type was determined in each site. Cultivars differing in maturity were simulated and their performance in the different environment types analysed. The simulations reproduced the ex- pected genotype–environment interactions between genotype matur- ity and stress pattern, thus indicating the potential of this method to unravel more complex trait–environment interactions. Simulations also provided insight into the key adaptive traits associated with improved performance in each environment type, with flowering date and the amount of water use pre- and post-anthesis having a major role. Using crop modelling for environmental characterisation can help breeding programs to set up their trials and select for germplasm best adapted to the targeted regions. Email Address for correspondence: karine.chenu@dpi.qld.gov.au doi:10.1016/j.cbpa.2009.04.566 P6.29 Poster Session — Monday 29th June 2009 A new network model explains the evolution of plant-specific metabolic networks Zoran Nikoloski (Institute of Biochemistry and Biology University of Potsdam), Patrick May (Max-Planck Institute for Molecular Plant Physiology), Joachim Selbig (Max-Planck Institute for Molecular Plant Physiology) For a set of metabolic pathways, we define six models of metabolic networks including all possible pairwise relationships between meta- bolites, enzymes, and reactions. The unified study of these models provides an accurate systematic framework for evolutionary interpreta- tion of the salient network properties. Here, we analyze the network properties of plant-specific metabolic networks assembled from Biocyc for the following species: Arabidopsis thaliana, Capsicum anuum, Coffea canephora, Solanum lycopersicum, Solanum lycopersicum, Solanum tuberosum, Sorghum biolor, Medicago trunculata, Petunia hybrida, Nicotiana tabacum, and Oryza sativa. In addition, we compare and contrast the obtained results with those from: Escherichia coli, Synechocystis, Candida albicans, Saccharomyces cerevisiae, Dictyostelium discoideum, Chlamydomonas reinhardtii, Mus musculus, and Homo sapiens. As a result of our empirical study, we propose a probabilistic growth model which (i) can explain the evolutionary characteristics of metabolic networks, (ii) differs from the well-established scale-free network models, (iii) demonstrates scale-free behavior of the degree distribution in both tails. We argue that the theoretical and empirical analyses of the proposed model, in terms of few salient network invariants, can help identify the evolutionary role of enzymes and metabolites through their participation in reactions. Email Address for correspondence: nikoloski@mpimp-golm.mpg.de doi:10.1016/j.cbpa.2009.04.567 P6.31 Poster Session — Monday 29th June 2009 The dynamics of resource allocation during rice (Oryza sativa) seedling establishment Saritha Kappalla (University of Liverpool), Ghulam M. Subhani (Uni- versity of Agriculture Faisalabad), N.K. Sathymoorthy (Tamil Nadu Agricultural University), Meriel G. Jones (University of Liverpool), Martin Mortimer (University of Liverpool) Rice cultivar selection for improved seedling establishment and vigour is essential as direct seeding is increasing adopted to combat the water shortages faced in traditional rice transplanting. Seedling establishment in rice involves developmental morphogenesis from the embryo and the successive expression of a coleorhiza, coleoptile, radicle and plumule and subsequently root and shoot tissues. This transition from embryo to autotrophic seedling is highly sensitive to flooding induced stresses. We measured component changes during this transition and described the dynamics by non-linear regression. Endosperm utilization, prior to the onset of autotrophy, was found to vary noticeably across a cultivar range exhibiting adaptations to both saturated soil water environments (cvs IR72, IR64) and aerobic ones (cvs PSBRC09, Azucena). Cultivars differed in coleoptile responsiveness (extension rate and maximum length expressed) to timing and depth of flooding, as did the expression of root tissues. Whilst having similar initial seed weights, cultivars also differed in rates of endosperm mobilization and in proportional allocation of seed biomass to root and shoot. cv Azucena was able to sustain growth in darkness for a significantly longer duration than other cultivars, whilst being much more sensitive to early flooding. The transcriptomics of gene expression underlying these phenotypic responses is currently being investigated. Email Address for correspondence: kalle@liverpool.ac.uk doi:10.1016/j.cbpa.2009.04.569 P6.32 Poster Session — Monday 29th June 2009 Improving the genome-scale metabolic network of Arabidopsis thaliana Nils Christian (MPI of Molecular Plant Physiology), Oliver Ebenhöh (MPI of Molecular Plant Physiology), Patrick May (MPI of Molecular Plant Physiology) The development of modern high-throughput technologies has enabled biologists to collect an immense amount of data character- izing the state of a cell or organism. The present challenge of systems biology is now to integrate this diverse information to develop a systems wide understanding of biological processes. We present a strategy that incorporates genomic sequence data and metabolite profiles into modeling approaches to arrive at improved gene annotations and more complete genome-scale metabolic networks. Abstracts / Comparative Biochemistry and Physiology, Part A 153 (2009) S219–S229 S227