Morphological characterization of papaya accessions in Jember through multivariate analysis

Abstract

This study aimed to select the most informative morphological descriptors in characterizing and distinguishing papaya accessions in Jember. The multivariate analysis was used to i) determine the closeness between the papaya accessions collected based on morphology descriptors and ii) define a list of sufficient minimum descriptors to distinguish papaya accessions in Jember. Five quantitative and fifteen qualitative descriptors were used in evaluating 23 accessions. The data collected through quantitative descriptors were analyzed by the means of PCA­biplot analysis and qualitative descriptors were analyzed using entropy level. Statistical analysis was carried out by R studio software using two packages, i.e. biotools and factoextra packages. Based on the research results, the first two principal components of the PCA­biplot representation of 23 accessions were able to evaluate 67.18% of the total variability. The accessions located in KAL­SUK, GUM­PUR 1, and GUM­PUR 2 had similarities in tree diameter and width of mature leaf characters, while KAL­PLA, MUM­SUM 1, LED­SLA, MUM­LEN 1, SUM­RO 2, and SUM­JAM accessions shared similarities in the length of mature leaf, petiole, and peduncle. The minimum descriptors that could be used to characterize papaya plant accessions in Jember were two quantitative descriptors (tree diameter and length of mature leaf) and three qualitative descriptors (general shape of mature leaf teeth, height to first fruit, and stem color). The combination of quantitative­qualitative descriptors allowed the simple and rapid characterization of papaya accession. In addition, it could reduce costs and labor without losing information in the characterization of papaya plants.

Keywords

• pca­biplot analysis
• papaya
• variability
• variable selection