Genetic dissimilarity and selection of shoot morphological describers in jambu [Acmella oleracea (L.) R.K. Jansen] (Asteraceae)

Abstract

Genetic variability, often quantified using morphological descriptors, is fundamental to plant breeding programs as it underpins genetic gains. This study evaluated genetic diversity among 25 jambu (Acmella oleracea) populations, focusing on shoot-related traits. Plants were sourced from rural farms and open markets across seven municipalities in northeastern Pará, Brazil, and cultivated in a greenhouse at the Federal Rural University of the Amazon, Capitão Poço campus. Fifteen morphological descriptors were assessed using a completely randomized design with three replicates, each plot comprising two plants. Descriptor selection was performed via Singh’s method, canonical variable analysis, and phenotypic correlation. Genetic dissimilarity was quantified using canonical variables and hierarchical clustering. Singh’s method identified leaf blade length, fresh plant mass, and fresh stem mass as the least informative traits. Canonical variable analysis recommended excluding leaf blade width and fresh leaf mass, while phenotypic correlations indicated redundancy in fresh plant mass, fresh stem mass, and leaf blade length. Cluster analysis delineated four groups based on canonical variables and three via unweighted pair group method with arithmetic mean (UPGMA), with congruent groupings between methods. These results confirm significant genetic variability among jambu populations and demonstrate that discrimination can be achieved efficiently using a streamlined set of morphological descriptors. This work advances jambu breeding by proposing trait selection strategies that reduce evaluation costs without compromising genetic diversity assessments.