Navigating the Era of Neo-Heirloom Tomatoes (Solanum lycopersicum L.): Developing an Heirloom Market Type Classifier Using High-Throughput Phenomics and Machine Learning

Date

2022

Contributor

Instructor

Depositor

Speaker

Researcher

Consultant

Interviewer

Narrator

Transcriber

Annotator

Journal Title

Journal ISSN

Volume Title

Publisher

University of Hawaii at Manoa

Volume

Number/Issue

Starting Page

Ending Page

Alternative Title

Abstract

Although there are various informal definitions for heirloom tomatoes (Solanum lycopersicum L.), there is no official botanical classification or certification scheme. Novel hybrid cultivars advertised as 'heirloom-like' exemplify the term's application as a phenotypic descriptor, departing from the term's established conceptions. Two field trials were conducted in Hawai'i to screen cultivars representing traditional heirlooms, heirloom hybrids, and commercial cultivars to explore this concept. We observed significant differences in total and marketable yields (kg/plant) within and between three tomato market classes. Phenotypic traits related to fruit morphology, color, and physicochemical quality were used to develop a Multilayer Perceptron Neural Network (MLPNN) classifier to investigate whether the heirloom archetype could be defined. The model could distinguish between traditional heirlooms, heirloom hybrids, and commercial cultivars with an accuracy rate of 85%. Global and local agnostic tests identified traits for physicochemical quality, color, distal end shape, blockiness, and the latitudinal section as most influential for distinguishing the heirloom class. This study demonstrated that a wide range of phenotypic traits could be selected to target the heirloom ideotype; however, quality standards should be considered to preserve the integrity and value of the heirloom insignia.

Description

Keywords

Heirloom tomatos, Phenomics, Solanum lycopersicum

Citation

Extent

Format

Geographic Location

Time Period

Related To

Related To (URI)

Table of Contents

Rights

Rights Holder

Local Contexts

Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.