Comparative analysis of different methods revealing the involvement of proteins during ageing of rice seeds

Seed longevity is a very important characteristic controlling seed quality. However, the mechanism underlying this characteristic is poorly understood. In this study, 'FH7185', a storable rice variety, the germination rate of which was 66.63%, much higher than the control after artificial ageing under relative humidity 88% and 42°C for 21 days. Different methods were applied to reveal the involvement of proteins during ageing of rice seeds. In total, 35 differential proteins were identified from 2D‐PAGE, and 3,719 proteins from iTRAQ analysis. A comparison of these two methods showed that not all protein types could be detected on the 2D‐PAGE gels, and the dynamic range was somewhat limited. So the comprehensive proteome map from the iTRAQ analysis was used to identify the quantitatively regulated proteins that played roles in seed longevity, and found that the most marked change was the increased abundance of many metabolic enzymes, especially the ones involved in α‐linolenic acid metabolism in the embryos during ageing. OsLOX2 and OsLOX3 had a negative effect on seed longevity, which were lower in FH7185 than the control. The dehydrogenase (LOC_Os07g44430) which played a major catalytic role in lipid peroxidation, also changed significantly during ageing. Therefore, OsLOX2, OsLOX3 and Loc_OS 07G44430 may be involved in the regulation of seed longevity with a synergistic effect.     

采用卷积神经网络构建西北太平洋柔鱼渔场预报模型

对远洋渔场资源和位置进行预报可以为远洋渔业生产及管理提供重要信息。该研究针对西北太平洋柔鱼渔场，利用海洋表面温度遥感信息和中国远洋渔船生产资料，基于深度学习原理，选取卷积神经网络构建西北太平洋柔鱼渔场预报模型。根据不同月份、不同通道构建了多种数据集，用于训练渔场预报模型。训练结果表明，4个通道组合的数据集的训练结果最优，渔汛早期（7-8月）、中期（9月）和后期（10-11月）测试样本的准确率分别为80.5%、81.5%和81.4%。以2015年的真实渔场数据对模型进行验证，模型的平均召回率为82.3%，平均精确率为66.6%，F1得分平均值为73.1%，预测的高产渔区与实际作业的高单位捕捞努力量渔获量区基本匹配。该研究构建的渔场预报模型可以获得较好的准确率，可为其他鱼种的渔场预报模型构建提供新的思路。     

Deep neural network algorithm for estimating maize biomass based on simulated Sentinel 2A vegetation indices and leaf area index

Accurate estimation of biomass is necessary for evaluating crop growth and predicting crop yield.Biomass is also a key trait in increasing grain yield by crop breeding.The aims of this study were(i)to identify the best vegetation indices for estimating maize biomass,(ii)to investigate the relationship between biomass and leaf area index(LAI)at several growth stages,and(iii)to evaluate a biomass model using measured vegetation indices or simulated vegetation indices of Sentinel 2A and LAI using a deep neural network(DNN)algorithm.The results showed that biomass was associated with all vegetation indices.The three-band water index(TBWI)was the best vegetation index for estimating biomass and the corresponding R2,RMSE,and RRMSE were 0.76,2.84 t ha−1,and 38.22%respectively.LAI was highly correlated with biomass(R2=0.89,RMSE=2.27 t ha−1,and RRMSE=30.55%).Estimated biomass based on 15 hyperspectral vegetation indices was in a high agreement with measured biomass using the DNN algorithm(R2=0.83,RMSE=1.96 t ha−1,and RRMSE=26.43%).Biomass estimation accuracy was further increased when LAI was combined with the 15 vegetation indices(R2=0.91,RMSE=1.49 t ha−1,and RRMSE=20.05%).Relationships between the hyperspectral vegetation indices and biomass differed from relationships between simulated Sentinel 2A vegetation indices and biomass.Biomass estimation from the hyperspectral vegetation indices was more accurate than that from the simulated Sentinel 2A vegetation indices(R2=0.87,RMSE=1.84 t ha−1,and RRMSE=24.76%).The DNN algorithm was effective in improving the estimation accuracy of biomass.It provides a guideline for estimating biomass of maize using remote sensing technology and the DNN algorithm in this region.     

基于改进极限学习机的水体溶解氧预测方法

为了有效地指导水产养殖生产，提高溶解氧浓度预测的精度，提出了基于因子筛选和改进极限学习机（Extreme Learning Machine，ELM）的水产养殖溶解氧预测模型。首先，利用皮尔森相关系数法计算各影响因子与溶解氧浓度间的相关系数，提取强关联因子，降低预测模型的输入量维度；采用偏最小二乘算法（Partial Least Square, PLS）优化传统ELM神经网络，避免网络中隐含层共线性问题，保障输出权值的稳定性；然后，结合新型激活函数，构建水体溶解氧浓度预测模型。最后，将SPLS-ELM（Selection Based Partial Least Square Optimized Extreme Learning Machine）预测模型应用到江苏省无锡市南泉基地某试验池塘的水体溶解氧预测中。试验结果表明：该模型的预测均方根误差为0.3232，与最小二乘支持向量机（Least Square Support Vector Machine，LSSVM）、BP神经网络、粒子群（Particle Swarm Optimization，PSO）优化LSSVM和遗传算法（Genetic Algorithm, GA）优化BP神经网络相比分别降低40.98%、44.48%、34.73%和44.18%。且该模型的运行时间仅0.6231s，预测精度和运行效率明显优于其他模型。该模型的溶解氧预测曲线接近真实溶解氧变化曲线，能够满足水产养殖实际生产对水体溶解氧预测的要求。     

采用全卷积神经网络与Stacking算法的湿地分类方法

高精度湿地制图对湿地生态保护与精细管理具有重要的支撑作用。针对传统湿地分类方法的精度不高和泛化能力弱等问题，提出了一种联合全卷积神经网路（fully convolutional neural network，FCN）与集成学习的湿地分类方法。首先利用全卷积神经网络（SegNet、UNet及RefineNet）对GF-6影像的语义特征进行提取与融合，然后利用Stacking集成算法对融合后的特征进行判别和分类。结果表明，联合全卷积神经网络与Stacking算法能有效提取湿地信息，总体分类精度为88.16%，Kappa系数为0.85。与联合全卷积神经网络与单一机器学习RF、SVM与kNN算法相比，该文提出的湿地分类方法在总体分类精度上分别提高了4.87%，5.31%和5.08%；与联合单一全卷积神经网络（RefineNet、SegNet、UNet）与Stacking算法下的湿地分类结果，该文提出的湿地分类方法在总体分类精度上分别提高了2.78%，4.48%与4.91%；该文方法一方面能通过卷积神经网络提取遥感影像深层的语义特征，另一方面通过集成学习根据各分类器的表征性能进行合理的选择并重组，从而提高分类精度及其泛化能力。该方法能为湿地信息提取及土地覆盖分类方法的研究提供参考。     

花龄期棉花虫害的电子鼻检测

棉花害虫具有隐蔽性、迁飞性和突发性特点，并且影响因素众多，棉花虫害准确地诊断是农业领域的难点问题。该研究以受到棉铃虫侵害的花铃期棉花为研究对象，采用电子鼻对不同处理的棉花挥发物进行检测。研究表明，主成分分析（Principal Component Analysis, PCA）和聚类分析结果显示健康棉花释放的挥发物具有明显的昼夜节律性，健康棉花与虫害棉花差异性显著。径向基函数神经网络（Radial Basis Function Neural Network, RBFNN）对8个不同时间的4组虫害棉花处理进行分析，测试集判别总的正确率为73.4%，健康棉花对照组测试集判别正确率100%，误判样本出现在3个虫害处理之间。当不考虑时间因素建立虫害棉花统一的预测模型，RBFNN模型对健康棉花对照组的预测正确率均达到了100%，分析结果可以作为花铃期棉花是否遭受棉铃虫侵害的依据，说明电子鼻可以作为棉花虫害发生的有效监测手段，在农作物虫害监测领域具有潜在的应用价值。     

Artificial Diet Influences Population Growth of the Root Maggot Bradysia impatiens (Diptera: Sciaridae)

In order to investigate the effects of artificial diets on the population growth of root maggot Bradysia impatiens, its population growth parameters were assayed on eight artificial diets (Diet 1, D2, D3, D4, D5, D6, D7, and D8). Results showed that developmental duration from egg to pupa was successfully completed on all eight artificial diets. However, the egg to pupal duration was shortest, while the survival rate of four insect stages was lowest when B. impatiens was reared on D1. When B. impatiens was reared on D7 and D8, the survival rate, female longevity, and female oviposition were higher than those reared on other diets. When B. impatiens was reared on D7, the intrinsic rate of increase (r_m = 0.19/d), net reproductive rate (R₀ = 39.88 offspring per individual), and finite rate of increase (λ = 1.21/d) were higher for its population growth with shorter generation time (T = 19.49 d) and doubling time (Dt = 3.67 d). The findings indicate that the D7 artificial diet is more appropriate for the biological parameters of B. impatiens and can be used an indoor breeding food for population expansion as well as further research. We propose that vitamin C supplement added to the D7 is critical for the improvement of the B. impatiens growth.     

基于广义回归神经网络的灌溉系统首部多用户配水快速PID控制

针对多用户配水状态下灌区流量、压力需求变化范围大,传统流量、压力控制响应速度慢等问题,建立适用于多用户灌区配水的灌溉系统首部控制技术。该研究通过分析供水系统流量、压力调节方式,提出了流量、压力PID(Proportion Integration Differentiation)耦合调节方法,建立以电动阀开度为流量控制量、变频器频率为压力控制量对流量和压力进行调控的灌溉首部控制系统。为了减少系统的调节时间,提高系统的运行效率,采用广义回归神经网络(Generalized Regression Neural Network,GRNN)建立流量、压力和电动阀控制模拟量、变频器控制模拟量间的预测模型,形成神经网络PID控制模型(GRNN_PID),并进行模型精度和控制精度验证。GRNN训练结果显示,变频器控制模拟量的相对误差为0.11%～3.86%,电动阀控制模拟量相对误差为0.09%～5.74%,模型精度较高。使用3个调节过程模拟3个用户的需水行为对模型进行验证,结果表明,GRNN_PID模型3个过程的调节时间分别为11.6、10.7和7.2 s,PID模型3个过程的调节时间分别为31.7、29.6和16.9 s,GRNN_PID模型大幅减少了系统的调节时间,提高了系统的运行效率;分别计算了2种模型的控制精度,GRNN_PID调节方法和传统PID调节方法的稳态流量和压力误差都在1%以内,最大超调量为8%,控制精度较高但相差不大,表明GRNN是从策略上加速系统调节速度,其本身并没有对PID的参数进行调整,因此对系统的控制精度影响不大。研究可为灌溉系统流量压力快速控制提供参考。