Correlation Analysis Of Main Cotton Yield Characters And Molecular Marker SSR In A Synthetic Population

Upland cotton is not only a cardinal yield crop, but also the largest natural fiber crop in the world. It is urgent to improve the cotton yield as the improvement of living standard and the development of textile industry.We make some molecular markers analysis by a synthetic population constructed from 16 parents' hybridization with the correlation analysis methods of t-test and variance analysis, and the molecular marker SSR as technology support. The purpose of our study is to find out the genetic loci associated with cotton yield characters. The main results are as follows.1. The analysis of genetic parameters and correlation analysis between generations of main yield characters indicate that the broad sense heritability of lint percentage is higher than any other trait, with the value of 64.52%, the correlation of which is also the highest, with the value of 0.672. The variation coefficients of plots, genetic variation coefficient and 5% genetic advance of lint percentage are lower with the value of about 5%, which of lint cotton yield, seed cotton yield and boll number per plant are moderate while the broad sense heritability and correlations between them stand lower. The foregoing results show that the influence of environment on lint percentage is the lowest and the heredity between generations of lint percentage is of the most steadinesses, while the variation extent and effect of choice are low. The influence of environment on lint cotton yield, seed cotton yield and boll number per plant is higher and the heredity between generations of these is unsteady, while the variation extent and effect of choice of these are higher.2. The correlation analysis between the yield characters indicates that both the phenotypic and genetic correlation coefficients between seed cotton yield and boll number per plant, single boll weight are significant difference at 0.01 level, the phenotypic and genetic correlation coefficients between seed cotton yield and boll number per plant, single boll weight, lint cotton yield are also significant difference at 0.01 level. The foregoing results show that boll number per plant and single boll weight are significance to the yield. The comparison of the phenotypic and genetic correlation coefficients shows that the correlations between the yield characters of cotton are influenced by environment obviously.3. The correlative analysis of the molecular markers and yield characters indicates that the markers with the highest positive effect on lint percentage is BNL1065-B, the relative effect of which is 4.043%; the markers with the highest positive effect on single boll weight is TMG21-B, the relative effect of which is 4.770%, the markers with the highest positive effect on boll number per plant is BNL3261-A, the relative effect of which is 9.534%, the markers with the highest positive effect on seed cotton yield is BNL3261-A, the relative effect of which is 11.654%, the markers with the highest positive effect on lint cotton yield is BNL3261-A, the relative effect of which is 12.382%. There also appear a lot of markers with negative effect on the characters. For each character, the correlative marker with effect of the biggest absolute value is negative. The higher the coherence of the result of t-test and variance analysis is, the higher the variation extent of characters is. It is more effective of variance analysis than t-test in the correlative analysis of the molecular markers and yield characters.4. The analysis of interaction effect between molecular markers indicates that there are remarkable but not determinate interaction effects between markers. So we have to ensure that the expected effects could be reached before the molecular marker-assisted selection with two markers together is carried out to improve a certain yield character of cotton. A simple and effective method is to compartmentalize the data into 4 groups by the two markers, compare the effects of each group and choose the best one.5. The analysis of molecular markers and their correlative yield characters indicates that there are 20 markers correlated to more than one yield characters, the effects to its correlative yield characters of each marker of these are all positive or all negative, which is coincident with the result that the correlation coefficients between yield characters are all positive. The phonotype correlations between yield characters are coincident with the interrelation between yield characters at molecular level by molecular markers. The significance of boll number per plant is highest to the yield, single boll weight is less significant, and lint percentage is of the lowest significance. There are no markers correlated to boll number per plant, single boll weight and lint percentage at the same time, only 8 markers are correlated to two of the 3 characters and only 4 markers of them have positive effect value.