| Maize(Zea mays L.)is an annual Grass of the corn of Gramineae.It has a strong adaptability.It was used as a “rescue plant” when it was transmitted to China.Maize has now become one of the most important economic crops and it has a very important position in the international food market.In China,maize is an important grain-feeding crop with a wide planting area and high yield,which plays an indispensable role in life and production.In recent years,the Northern Corn Leaf Blight(NCLB)has frequently erupted,which has a serious impact on the yield of maize.Northern Corn Leaf Blight is one of the major diseases in the growth of maize caused by Setosphaeria turcica It is a quantitative trait controlled by a few major genes and a large number of minor genes.Genomic Selection(GS)breeding technology can effectively predict the complex quantitative traits of micro-effect and multi-gene control,and has been gradually applied to effective molecular breeding methods in plants.The measure of efficiency is the prediction accuracy of genome selection,which is the correlation coefficient between the phenotypic values determined in the field and the genomic estimated breeding values obtained from the whole-base genome prediction.In this study,three double haploid populations were used to research the accuracy of genome selection for the Northern Corn Leaf Blight.Use the software Icimapping to delete redundant markers in order to get Bin markers.The genetic map was constructed by Bin marker to locate the QTL of the Northern Corn Leaf Blight.And detecting the significance of the Bin markers and obtaining the effect value of the Bin markers.The Bin marker was used as a fixed effect marker to analyze its influence on the prediction accuracy of genome selection,and then the modeling scheme was optimized to improve prediction accuracy and provide reference for actual breeding work.Applying Bin markers to establish a Genome selection prediction Model.The results are as followings:(1)The prediction accuracy with better results can be obtained by modeling and predicting with 50% of the number of markers as the training population.(2)Putting the Bin markers into the training population can improve the efficiency of the genomic selection technique.The significance of the markers and its proportion together affect the prediction accuracy.The more bin markers in the training population,the higher prediction accuracy can be obtained.(3)Let all Bin markers as fixed markers plus random markers to 50% of all markers put into the prediction model,and the prediction effect is best.(4)In this study,using the top 60% of the significant Bin-marked complementary random markers for prediction can achieve higher prediction accuracy than the same number of random markers for GS prediction.(5)The three DH groups used all Bin markers add random markers to make the training population reach 50% of the total markers.Compared with the 50% markers of the total markers,the prediction accuracy of the Northern Corn Leaf Blight by modeling was significantly improved by 3.33%,12.8% and 10.22%,respectively.Compared with the total markers,the prediction accuracy of the Northern Corn Leaf Blight by modeling was significantly improved by 3.68%,13.2% and 10.38%,respectively.Therefore,in actual breeding,a small parent population with a small number of major and a large number of minor gene-controlled traits is known.And consider Bin markers combined with QTL analysis as a first step to integrate into the genome prediction program to optimize the breeding process and improve the prediction accuracy. |
PDF Full Text Request