Short Communication Comparison of efficiency of distance measurement methodologies in mango (Mangifera indica) progenies based on physicochemical descriptors

Abstract

ABSTRACT. We investigated seven distance measures in a set of observations of physicochemical variables of mango (Mangifera indica) submitted to multivariate analyses (distance, projection and grouping). To estimate the distance measurements, five mango progeny (total of 25 genotypes) were analyzed, using six fruit physicochemical descriptors (fruit weight, equatorial diameter, longitudinal diameter, total soluble solids in °Brix, total titratable acidity, and pH). The distance measurements were compared by the Spearman correlation test, projection in twodimensional space and grouping efficiency. The Spearman correlation coefficients between the seven distance measurements were, except for the Mahalanobis' generalized distance (0.41 ≤ rs ≤ 0.63), high and significant (rs ≥ 0.91; P < 0.001). Regardless of the origin of the distance matrix, the unweighted pair group method with arithmetic mean grouping method proved to be the most adequate. The various distance measurements and grouping methods gave different values for distortion (-116.5 ≤ D ≤ 74.5), cophenetic correlation (0.26 ≤ rc ≤ 0.76) and stress (-1.9 ≤ S ≤ 58.9). Choice of distance measurement and analysis methods influence the characterization of genetic variability, and this should be taken into account for studies of mango.

Genetic Breeding and Diversity of the Genus Passiflora: Progress and Perspectives in Molecular and Genetic Studies

“…Methodological influences on the results of diversity estimates have been reported for a variety of species, including Mangifera indica [99], Olea europaea [100], Zea mays L. [101], and Solanum lycopersicum [102]. Differences in methodologies have also been discussed for studies of morphological characteristics and of genetic diversity that used molecular markers to study passion fruit [103].…”

Short Communication Comparison of methods for estimates of molecular genetic diversity in genus Croton: influence of coefficients, clustering strategies and data projection

“…Similar to the results observed for hierarchical clustering methods and for projection in two-dimensional space, the formation of groups with grouping methods by optimization (Tocher method and Tocher modified method) showed variations when using different coefficients (data not shown). Variations in the results from clustering by optimization methods were also observed by Alves et al (2012) and Duarte et al (1999), demonstrating the importance of the choice of coefficients to be used in estimating similarity (or dissimilarity), since this choice influences the results of grouping, regardless of the method chosen for training groups. It should be noted that the combined use of hierarchical clustering methods and optimization is commonly observed in the literature, for example, in the articles of Duarte et al (1999) with common bean, Bertan et al (1999) with wheat genotypes, and Alves et al (2012) with progeny of mango.…”

“…Scaldaferri et al 0.57) and Ochiai (r c = 0.55) coefficients. Contrasting results in relation to the projection of the genetic distances in two-dimensional are available in the literature, such as the high-stress levels of these projections in studies with passion fruit (54 < S < 75), based on RAPD markers (Cerqueira-Silva et al, 2009), and lower values in studies with common bean (Phaseolus vulgaris L.) (11 < S < 57), based on RAPD markers (Duarte et al, 1999), and with mango (Mangifera indica L.) (17 < S < 23), based on fruit physicochemical descriptors (Alves et al, 2012 The different combinations between 10 coefficients and the 8 hierarchical clustering methods generated distinct results concerning the efficacy of the grouping data in presenting the original distance data (-32370 ≤ D ≤ 55.1; 0.10 ≤ r c ≤ 0.77; 5.9 ≤ S ≤ 1703) (Table 4). In summary, the methods WPGMA (0.41 < r c < 0.75; -26.19 < D < 2.52; 6.3 < S < 18.95) and UPGMA (0.41 < r c < 0.77; 0.3 < D < 1.5; 5.99 < S < 12.61) were the ones that showed the best results for C. linearifolius.…”

“…Among the statistical methods used for diversity studies are i) similarity (and dissimilarity) coefficients, ii) hierarchical grouping methods and grouping methods by optimization, and iii) data projection in two-dimensional space (Duarte et al, 1999;Mohammadi and Prasanna, 2003). In view of the large number of statistical methods available in the literature and the fact that existing comparative studies reveal that the choice of statistical methods can affect the results (Meyer et al, 2004;Gonçalves et al, 2008;Cerqueira-Silva et al, 2009;Sesli and Yegenoglu, 2010;Alves et al, 2012), it is important to establish criteria for choosing statistical methods that are appropriate for the reality of each study to be conducted. M.M.…”