Combining quantitative and qualitative descriptors to predict genetic diversity in Capsicum

Abstract

The objective of the present study was to assess the accuracy of analyzing quantitative and qualitative descriptors separately or combined to differentiate Capsicum accessions. We assessed 47 Capsicum accessions from the UFV Vegetable Germplasm Bank (BGH-UFV) with botanical classification previously known. The experiment was arranged in a randomized block design with four replicates. Variables consisted of 16 morphological descriptors proposed by the IPGRI. Analysis of quantitative descriptors alone was performed by calculating the Mahalanobis distance matrix (D1), while qualitative descriptors alone were analyzed using the simple coincidence index (D2). The joint analysis consisted of the sum of the distance matrices D1 and D2, and the joint calculation of both descriptors using Gower’s algorithm. Association between the distance matrices was assessed by Mantel’s correlation test, and accession clustering was performed using the UPGMA method. There was genetic variability amongst Capsicum accessions for both quantitative and qualitative traits. Accession clustering was more accurate when performed using the distances obtained by the simple coincidence index (qualitative descriptors alone) and the Gower’s algorithm (qualitative and quantitative descriptors combined)