THE GENETIC VARIATION BETWEEN MORPHOLOGICAL AND MOLECULAR MARKERS IN CUCURBITA PEPO L. BY USING MULTIPLE CORRELATION AND PCA

Abstract

A multiple correlation was used between 15 inbreds of Cucurbita pepo L., depending on the results of morphological markers, and then described clustering by using principal component analysis. It was found that M, C and F inbreds were with a low correlation, while the other inbreds associated with a highest correlation to each other of (r = 0.87**) .Result of PCA (based on morphological markers) divided to several clusters. The first cluster included the number of branches, number of female flowers, number of fruits per plant, yield per plant and the sex ratio. The second cluster included plant height, number of leafs and number of male flowers. The third cluster included the date of first harvest, fruit weight and number of nodes and the fourth cluster included the length and shape of the fruits, while we find not correlation to each of the markers of fruit diameter and the ratio of fruit set. The result of the PCA (based on the results of molecular markers) absence correlation A and B pure lines with others, while the others inbreds correlate with each other within one cluster divided several clusters. The first cluster included C and D pure lines. The second cluster included K, N and O pure lines. The third cluster included E, F, G, H and J inbreds. It was found that the correlation coefficient between the morphological markers and molecular markers of low correlation and non significant (r = 0.02), which shows a low correlation between cluster analysis of molecular markers and cluster analysis of morphological markers. Accordingly, finding correlation coefficients and principal component analysis were useful to have a comprehensive understanding of genetic variations among inbreds before crossing program.