| As a typical perennial shrub plant,cassava is widely used in food and feed.Due to it has many characteristics such as high biomass,resistance to barrenness,diseases,pests and drought.It is also widely used in production and processing,such as paper making,textile,starch production and biofuel production.Although the current cassava breeding system is relatively mature,the traditional breeding methods can no longer meet the needs of production and life.Because the breeding period of traditional breeding is long.And cassava genome has high heterozygosity,thus increases the abundance of the genetic variation of cassava,but also increases the difficulty of the cassava breeding.So the use of molecular marker-assisted breeding to research alleles that control excellent traits is of great significance for the genetic improvement of cassava.The experimental population in this study was came from 192 cultivated cassava varieties in different place of Brazil.A total of 6 quantitative traits including plant height,aerial part yield,fresh root yield,harvest index,dry matter content,starch yield were investigated.The phenotype identification of12 different environments was carried out,and the phenotype observation values were measured and recorded.Subsequently,the phenotype traits was analyzed,and the best linear unbiased prediction method(BLUP)was used to estimate the breeding values of different traits.Then the generalized heritability(H2)was estimated by using the data generated during the BLUP calculation process.At the same time,we extracted the cassava genomic DNA of the experimental population,and used AFSM(the amplified fragment single nucleotide polymorphism and methylation)technology to simplified sequencing of DNA.Through statistical analysis,the whole genome level of SNPs and In Dels molecular markers of cassava population were obtained.The filtered high-quality molecular markers were used for genetic diversity analysis,population genetic structure analysis and genome-wide association analysis.Finally identify the significant loci and determine the candidate genes.The research results are as follows:After statistical analysis of the cassava phenotypic traits,it was found that each of the above six traits showed varying degrees of variation.Among them,the highest coefficient of variation was starch yield,and the range of the coefficient of variation was 0.32 to 1.04.The lowest coefficient of variation is the dry matter content,and the coefficient of variation ranges from 0.05 to 0.19.Subsequently,the generalized heritability of each trait was evaluated.The results showed that: the generalized heritability of dry matter content was the highest(0.89),the aerial part yield was the lowest(0.71).And the generalized heritability of the six traits were between 0.71 and 0.89.The results indicate that the six quantitative traits in the study have high generalized heritability,which means that they are relatively stable in different environments.We used 9,943 high-quality SNPs and In Dels molecular markers for analyzing.The group structure analysis showed that the 192 Brazilian cassava can divided into 9 subgroups,which is roughly the same as the result of principal component analysis and the clustering results of evolutionary tree.Simultaneously we calculated the genetic diversity index(?)of the cassava population and each subgroup.The genetic diversity index of this group is 0.274,and the genetic diversity index of subgroup 1 to subgroup 9 is between 0.192 and 0.289.The results showed that the Brazilian cultivated cassava population has a relatively high level of genetic diversity;after calculating the population differentiation index(Fst),it is found that except for the strong genetic differentiation between subgroup 1 and subgroup 4,all subgroups were moderately or weakly genetic differentiation,and the genetic differentiation index between subgroups is between 0.031 and 0.152.In addition,we also conducted a genetic distance analysis of the cassava population,and found that the genetic distance between these cassava germplasms was between 0.084 and0.297,with an average genetic distance of 0.228.After performing genome-wide association analysis(GWAS),the following results were obtained: a total of 196 significant loci were obtained under 12 environments,and 96 candidate genes were identified.Among the plant type traits(plant height),33 significant loci and 13 candidate genes were found by association analysis;among the yield-related traits(including fresh root yield,aerial part yield and harvest index),respectively 28,32,and 20 significant loci were detected,and 21,14,and 11 candidate genes were identified respectively;in root quality traits(including dry matter content and starch yield),31 and 52 significant loci were detected respectively,14 and 23 candidate genes were identified respectively.According to statistics,a total of 9significant loci have the same association in different traits.GO annotation analysis was used 75 candidate genes which obtained by annotating yield-related traits and removing duplicates.The annotation results showed that 10 candidate genes were classified into cellular component,mainly concentrated on the functions of cell,cell part,organelle,membrane.37 candidates Genes are classified into molecular function,mainly focused on catalytic activity and binding function.25 candidate genes are classified into biological process,mainly focused on the regulation of biological process,and cellular process. |
PDF Full Text Request