Identification Of Winter Wheat Spike Number And Accumulation And Partitioning Characteristics Of Photosynthetic Product

Genetic research on wheat?Triticum aestivum L.?physiology and yield phenotype have an important guiding value for the improvement of new cultivars with high and stable yield.The objectives of the present study were to analyze the dry matter accumulation and partitioning characteristics of wheat cultivars,and to explore the main effective quantitative trait loci?QTL?of spike number,two types of research were set up in this study.Aim I:seven elite cultivars?Jingdong 8,Lunxuan 987,Jingdong 17,Nongda 211,Zhongmai 175,Zhongmai 816 and Zhongmai 1062?from Northern Winter Wheat Zone in the past 20 years were selected.The cultivars dynamic dry matter accumulation and transportation in different organs during grain-filling,harvest index?HI?,and grain yield were investigated and proposed the cultivars improvement goals.Aim II:the way of machine learning was used to establish a wheat spike number validation model,which was trained and verified by large samples to improve the high-accuracy spike number intelligent identification system.Compared with the QTL results of wheat spike number per unit area under the methods of artificial ccount and intelligent identification,it confirmed the validity of the spike number identification system in genetic research and detected the main effective quantitative trait loci of wheat spike number per unit area.The main results are as follows:1.Zhongmai 816 and Zhongmai 175 had higher yield potential,which were 4923.0kg·hm^-2 and 4913.0 kg·hm^-2,respectively,mainly attributed to their higher biological yield and HI.The analyses of dry matter accumulation revealed that the photosynthetic area of the population at the seedling stage and the photosynthetic utilization efficiency during the grain filling stage had important effects on dry matter accumulation.Cultivars with low chlorophyll content,small flag leaf area?FLA?,and high photosynthetic rate?Pn?during the grain filling stage had higher dry matter accumulation efficiency and more soluble sugar storage in stems and leaves,such as Zhongmai 175.In addition,the photosynthetic compounds stored in Zhongmai 175 and Zhongmai 1062 before flowering had the highest contribution to grains,whereas Nongda 211 had the lowest.From the perspective of the dry matter distribution in vegetative organs,the dry matter storage of vegetative organs in Zhongmai 816 and Zhongmai175 at the flowering stage were slightly different from that of other cultivars,but the dry matter transportation efficiency in stems and leaves at the maturity stage were higher,and the residues of stored matter were less.Improving the light energy utilization efficiency of cultivars was a conductive way to increase the genetic yield potential.In addition,it was necessary to pay more attention to the selection of dry matter accumulation before the flowering stage and the dry matter transportation efficiency at the grain filling stage,reducing the residual photosynthetic products in tissues and organs at maturity,and improve the efficiency of substance transportation to grains.2.This study was identified 3 QTLs of the wheat spike number per unit area on 4DS,7DS,and 7DL,designated as QSnyz.caas-4DS,QSnyz.caas-7DS,and QSnyz.caas-7DL.Among them,MSN,ISN and VSN simultaneously located the QSnyz.caas-7DS on the 7D chromosome.3.The identification accuracy of wheat spike number under the method of R-CNN model was between 76%and 98%,with an average accuracy of 86.7%.In terms of the correlation analysis,the R² value of VSN and ISN was higher than the MSN,they were 0.83 and 0.50,respectively.Based on the results of the three methods of MSN,ISN and VSN,they were concluded that the R-CNN model can not only achieve rapid acquisition of spike number per unit area,but also provide a reference to predict wheat yield under suitable planting density.In addition,it was an effective and high-throughput method to identify potential QTL for wheat spike number.