Modeling protein accumulation in rice grain

中国人. 蛋白质含量是稻米的重要品质指标。通过解析和综合水稻植株氮素积累和转运的动态规律及其与环境因子和基因型的定量关系，构建一个基于氮流生理过程的水稻籽粒蛋白质形成模拟模型，以期为水稻生产中籽粒蛋白质指标的动态预测和管理决策提供量化工具。以田间试验资料为基础，结合已有水稻生长模型，采用生理发育时间作为定量生育进程的尺度，通过解析和综合水稻植株氮素积累和转运动态的基本规律及其与环境因子和基因型的定量关系，构建花前植株氮素吸收与积累、花后氮素吸收与转运及籽粒蛋白质形成过程的模拟模型。利用不同年份、不同生态点、不同品种类型和不同肥水条件下的大田试验资料对籽粒蛋白质形成模型进行了检验，籽粒蛋白质含量模拟值与观测值之间的决定系数大于0.84，根均方差（RMSE）小于0.26%。表明模型具有较好的通用性和可靠性，可以较准确地预测不同条件下的水稻籽粒蛋白质含量与蛋白质产量。基于植株氮素积累和转运的生理生态过程，以生理发育时间为主线，建立了较为简化的水稻籽粒蛋白质积累动态的模拟模型，模型的研究不仅为定量预测不同生态与肥水条件下不同水稻品种籽粒蛋白质含量与蛋白质产量的动态变化奠定了基础，而且是对国内外现有水稻生长与产量模拟模型的发展和完善。

英语. Protein content of grain is an important quality index of rice product. Analyzing the dynamic rule of nitrogen assimilation and translocation in rice plant and relationship, this study will development a simulation model on formation of grain protein based on the development process to predict rice grain protein content and yield under different ecological environmental conditions. On the basis of the field experiments with different cultivars and nitrogen levels, using physiological development time (PDT) as general time scale of development progress, the fundamental dynamic relationships between plant nitrogen accumulation and translocation and environmental and genetic factors were quantified and a simplified and explanatory simulation model was established to predict the processes of nitrogen uptake and partitioning in plant and protein formation in grain of rice by integrating with a rice growth model (RiceGrow). The model was validated using the data sets of different years, eco-sites, varieties, N fertilization and irrigation conditions with the RMSE 0.22%-0.26%. The results indicate that the model is accurate and applicable for predicting grain protein content and grain protein yield of rice under various environments. Using PDT as general time scale of development progress, this study developed the simulation model on rice grain protein formation based on nitrogen assimilation and translocation, which might predict rice grain protein content and yield under different ecological environment conditions, and expanded and perfected the exist rice growth models on quality prediction.