THE COMPARISON ANALYSIS OF SOFTWARE FOR MANTEL TEST BETWEEN DNA MARKERS AND MORPHOLOGICAL TRAITS OF PLANT VARIETIES

Abstract

An advanced approach to the assessment of varieties for the determination of their differences both within the DUS test and breeding is a combination of morphological traits and DNA markers. An implementation of this approach includes a correlation assessment between genetic distances matrices. To this end, the Mantel test is applied. The purpose of the study was to identify the main advantages and disadvantages of different software products for the Mantel test based on correlation investigation between DNA markers and morphological traits of lettuce varieties and maize lines. As a result of correlation calculation between 8 SSR markers and 36 morphological traits of 100 maize lines by XLSTAT (software for Microsoft Excel) p-value (probability of obtaining test results) was 0.0005. The value of this indicator obtained by PASSaGE software was 0.034. There was a pvalue of 0.045, which was calculated by GenAlEx 6.5 in Microsoft Excel (MS Excel). A similar result (0.036) was obtained by software environment R. The pvalues, which were calculated between 7 EST-SSR markers and 32 morphological traits for four lettuce varieties by XLSTAT, PASSaGE, GenAlEx, and R, were 0.033, 0.039, 0.038, and 0.035, respectively. In the study, the upper-tailed test served an alternative hypothesis type, the level of significance α was 0.05, the type of correlation was Pearson correlation, and the Monte Carlo method was used for p-value computation. Thus, the obtained p-values allow to reject the null
hypothesis (H0) and adopt the alternative hypothesis Ha of correlation (p < α). The correlation coefficient for maize lines was 0.05 and for lettuce varieties 0.65. Therefore, XLSTAT and software environment R are the most suitable instruments for correlation assessment between genetic distances.

Keywords: correlation coefficient, maize, lettuce, Mantel, p-value.