Stepwise regression analysis of the correlation between leaf traits and SSR markers of Malus sieversii

摘要

In this study, variation of leaf traits of 18 clones selected from 600 Malus sieversii clones collected from Tianshan Mountain, Xinjiang (China) was investigated. SSR molecular markers were prepared; correlation between leaf traits and SSR(Simple Sequence Repeats) marker was determined through stepwise regression. The results showed that there were significant differences between the 15 leaf traits. The tip angle α showed the largest average variation coefficient of 25.39%, whereas the leaf index L 1 /L 3 yielded the lowest variation of 10.31%. The repeatability of most leaf traits were more than 0.80, which indicated that genetic factors accounted for a high proportion of phenotypic leaf traits. The clones were completely separated with the clustered result of 15 leaf traits and the genetic distance were varied between 1.293 and 7.235. Furthermore, 18 clones with 30 pairs of polymorphic SSR primers, which were evenly distributed among 17 chromosome linkage groups, were subjected to cluster analysis; the clones were completely separated. The genetic distances varied between 0.089 and 0.689, with the average genetic distance was 0.4328. Cluster results of all leaf traits were not correlated with that of SSR markers. Stepwise regression results showed that the same leaf trait was closely related to several sites; likewise, the same site was closely related to several leaf traits. The sites correlation with each single trait varied between 11 and 20, in which 11 sites yielded a leaf index of L 1 /A 1 , 11 sites exhibited a leaf index of L 1 /A 4 , and 20 SSR sites were correlated with the width of 3/4 of the leaf (A 4 ). Therefore, phenotypic traits of the leaf were closely correlated with several SSR sites.