This study addresses the limitations in evaluating interspecific hybrids, which often rely on semi-subjective classical genebank descriptors with qualitative traits. The research employs recognized phenomic traits to quantify descriptors, facilitating the identification of phenomic traits that reflect the percentage of parental characteristics retained in interspecific hybrids. The study involved P. vulgaris (G5773 – CIAT genebank accession number) , P. acutifolius (G40001), P. parvifolius - P. montanus (G40102) accessions and interspecific cross hybrid line INB 47 (G52443), cultivated using CIAT seed regeneration procedures in a mesh house.

The methodology involved characterizing and classifying three accessions and a derived breeding line resulting from interspecific crosses, utilizing multivariate and machine learning techniques. The proportions of phenomic traits in the interspecific hybrid (line INB 47) relative to its parent accessions were determined through random forest analysis and a corresponding confusion matrix.

The results encompass the evaluation of seed and pod morphometric traits, physiological behavior, and yield performance. Notably, phenomic descriptors with the highest predictive power included Fm', Fo', Fs', LTD, Chl, seed area, seed height, seed Major, seed MinFeret, seed Minor, pod AR, pod Feret, pod round, pod solidity, pod area, pod major, pod seed weight, and pod weight. Physiological traits of the interspecific hybrid showed 2.2% similarity with P. acutifolius and 1% with P. parvifolius. Additionally, in terms of seed morphometric characteristics, the hybrid exhibited 4.5% similarity with P. acutifolius.

The study's conclusions highlight the successful determination of phenomic proportions of parental traits in interspecific hybrids. The methodology has the potential to: i) confirm the transfer of traits of interest from P. acutifolius and P. parvifolius into their hybrids; ii) establish selected traits as "phenomic markers" for preserving desired physiological traits from exotic parental accessions, while retaining key seed characteristics of elite common bean accessions; and iii) propose a quantitative tool to aid genebank curators and breeders in making well-informed decisions based on quantitative analysis.