为了弥补一次性建模分析的缺陷,提高小麦条锈病遥感监测模型的运行效率和精度,根据模型集群分析(Model population analysis,MPA)算法的特点,综合利用光谱区间选择算法和光谱点选择算法的优势,提出了一种联合相关系数(Correlation coeff...为了弥补一次性建模分析的缺陷,提高小麦条锈病遥感监测模型的运行效率和精度,根据模型集群分析(Model population analysis,MPA)算法的特点,综合利用光谱区间选择算法和光谱点选择算法的优势,提出了一种联合相关系数(Correlation coefficient,CC)与MPA的特征变量优选算法。在利用CC算法对全波段光谱进行特征变量选择的基础上,分别利用基于MPA思想开发的竞争性自适应重加权采样法(Competitive adaptive reweighted sampling,CARS)和变量组合集群分析法(Variable combination population analysis,VCPA)进一步优选对小麦条锈病敏感的特征变量,并利用偏最小二乘回归(Partial least squares regression,PLSR)算法构建了小麦条锈病遥感监测的CC-CARS和CC-VCPA模型。结果表明:联合CC MPA算法优选的特征变量构建的CC-CARS和CC-VCPA模型精度均高于CC、CARS和VCPA算法。3组验证集样本中,CC-CARS模型预测病情指数(Disease index,DI)与实测DI间的R^(2)_(V)较CC模型和CARS模型至少分别提高了6.78%和6.66%,RMSEV至少分别降低了15.31%和10.98%,RPD至少分别提高了18.08%和12.34%。CC VCPA模型预测DI与实测DI间的R^(2)_(V)较CC模型和VCPA模型至少分别提高了9.58%和0.73%,RMSEV至少分别降低了20.78%和3.86%,RPD至少分别提高了26.22%和4.02%。基于CC-MPA的光谱特征优选算法是一种有效的特征选择方法,尤其是利用CC-VCPA方法选择的特征变量数更少,模型预测效果更好,研究结果对光谱特征优选及提高作物病害遥感监测精度具有重要的参考价值。展开更多
变量选择方法可以实现对高维数据的降维,降低标定模型的复杂度以及提高模型的预测能力和可解释性,对建立高效可靠的预测模型具有重要意义。本文将模型种群分析(Model Population Analysis,MPA)用于近红外光谱标定建模过程的变量选择,结...变量选择方法可以实现对高维数据的降维,降低标定模型的复杂度以及提高模型的预测能力和可解释性,对建立高效可靠的预测模型具有重要意义。本文将模型种群分析(Model Population Analysis,MPA)用于近红外光谱标定建模过程的变量选择,结合MPA在同一空间反复抽取子集的特点,提出一种子集索引重用核-偏最小二乘(Subset Index Reuse Kernel-Partial Least Squares,SIRK-PLS)融合建模方法。该方法通过对预先计算的协方差矩阵进行索引,从本质上避免MPA框架下变量选择子集交叉验证和回归系数求解过程中的冗余计算,提高建模效率。此外,SIRK-PLS建模方法可以根据样本数和变量数的比例,实现建模算法的自动最优切换。通过标称近红外光谱玉米数据集对算法性能进行验证。结果表明,本文提出的SIRK-PLS建模方法收敛速度快、精度高,适用于移动红外光谱设备的自动快速降维建模,具有一定的应用前景。展开更多
本文收集了环烷烃类、环烯烃类、酮类、胺类、醚类、酯类等有机物在固定相角鲨烷和SE-30上的气相色谱保留指数,并采用基于Monte Carlo采样的模型集群分析（Monte Carlo sampling model population analysis,MCS MPA）方法进行了定量结构...本文收集了环烷烃类、环烯烃类、酮类、胺类、醚类、酯类等有机物在固定相角鲨烷和SE-30上的气相色谱保留指数,并采用基于Monte Carlo采样的模型集群分析（Monte Carlo sampling model population analysis,MCS MPA）方法进行了定量结构-色谱保留指数相关关系建模方法的比较研究。对于两种固定相上的有机化合物,分别采用不同的分子描述符予以表征,分子描述符的选择基于统计学与遗传算法。采用的建模方法包括多元线性回归（multivariate linear regression,MLR）、支持向量机回归（support vector machine,SVM）、径向基函数人工神经网络方法（radial basis function artificial neural networks,RBF ANN）,通过所建模型预测了独立外部测试样本的气相色谱保留指数。研究结果表明,对于本文所研究的数据,SVM回归方法的建模效果优于MLR与RBF ANN方法。展开更多
Dissecting the genetic architecture of complex traits is an ongoing challenge for geneticists.Two complementary approaches for genetic mapping,linkage mapping and association mapping have led to successful dissection ...Dissecting the genetic architecture of complex traits is an ongoing challenge for geneticists.Two complementary approaches for genetic mapping,linkage mapping and association mapping have led to successful dissection of complex traits in many crop species.Both of these methods detect quantitative trait loci(QTL) by identifying marker–trait associations,and the only fundamental difference between them is that between mapping populations,which directly determine mapping resolution and power.Based on this difference,we first summarize in this review the advances and limitations of family-based mapping and natural population-based mapping instead of linkage mapping and association mapping.We then describe statistical methods used for improving detection power and computational speed and outline emerging areas such as large-scale meta-analysis for genetic mapping in crops.In the era of next-generation sequencing,there has arisen an urgent need for proper population design,advanced statistical strategies,and precision phenotyping to fully exploit high-throughput genotyping.展开更多
文摘变量选择方法可以实现对高维数据的降维,降低标定模型的复杂度以及提高模型的预测能力和可解释性,对建立高效可靠的预测模型具有重要意义。本文将模型种群分析(Model Population Analysis,MPA)用于近红外光谱标定建模过程的变量选择,结合MPA在同一空间反复抽取子集的特点,提出一种子集索引重用核-偏最小二乘(Subset Index Reuse Kernel-Partial Least Squares,SIRK-PLS)融合建模方法。该方法通过对预先计算的协方差矩阵进行索引,从本质上避免MPA框架下变量选择子集交叉验证和回归系数求解过程中的冗余计算,提高建模效率。此外,SIRK-PLS建模方法可以根据样本数和变量数的比例,实现建模算法的自动最优切换。通过标称近红外光谱玉米数据集对算法性能进行验证。结果表明,本文提出的SIRK-PLS建模方法收敛速度快、精度高,适用于移动红外光谱设备的自动快速降维建模,具有一定的应用前景。
文摘本文收集了环烷烃类、环烯烃类、酮类、胺类、醚类、酯类等有机物在固定相角鲨烷和SE-30上的气相色谱保留指数,并采用基于Monte Carlo采样的模型集群分析（Monte Carlo sampling model population analysis,MCS MPA）方法进行了定量结构-色谱保留指数相关关系建模方法的比较研究。对于两种固定相上的有机化合物,分别采用不同的分子描述符予以表征,分子描述符的选择基于统计学与遗传算法。采用的建模方法包括多元线性回归（multivariate linear regression,MLR）、支持向量机回归（support vector machine,SVM）、径向基函数人工神经网络方法（radial basis function artificial neural networks,RBF ANN）,通过所建模型预测了独立外部测试样本的气相色谱保留指数。研究结果表明,对于本文所研究的数据,SVM回归方法的建模效果优于MLR与RBF ANN方法。
基金supported by the Priority Academic Program Development of Jiangsu Higher Education Institutionthe National Natural Science Foundation of China(Nos.91535103,31391632,and 31200943)+4 种基金the National High Technology Research and Development Program of China(No.2014AA10A601-5)the Natural Science Foundation of Jiangsu Province(No.BK2012261)the Natural Science Foundation of Jiangsu Higher Education Institution(No.14KJA210005)the Postgraduate Research and Innovation Project in Jiangsu Province(No.KYLX151368)the Innovative Research Team of University in Jiangsu Province
文摘Dissecting the genetic architecture of complex traits is an ongoing challenge for geneticists.Two complementary approaches for genetic mapping,linkage mapping and association mapping have led to successful dissection of complex traits in many crop species.Both of these methods detect quantitative trait loci(QTL) by identifying marker–trait associations,and the only fundamental difference between them is that between mapping populations,which directly determine mapping resolution and power.Based on this difference,we first summarize in this review the advances and limitations of family-based mapping and natural population-based mapping instead of linkage mapping and association mapping.We then describe statistical methods used for improving detection power and computational speed and outline emerging areas such as large-scale meta-analysis for genetic mapping in crops.In the era of next-generation sequencing,there has arisen an urgent need for proper population design,advanced statistical strategies,and precision phenotyping to fully exploit high-throughput genotyping.
