Discovering the "digital identity" of peanut seeds and seedlings: a technological step to mitigate varietal mixing in the post-harvest stage

• Grant number: 25/01449-9
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: June 01, 2025
• End date: May 31, 2027
• Field of knowledge: Agronomical Sciences - Agronomy - Crop Science
• Principal Investigator: Edvaldo Aparecido Amaral da Silva
• Grantee: Gustavo Roberto Fonseca de Oliveira
• Host Institution: Faculdade de Ciências Agronômicas (FCA). Universidade Estadual Paulista (UNESP). Campus de Botucatu. Botucatu , SP, Brazil

Abstract

One of the main sources of varietal mixture in the peanut seed production system occurs during the post-harvest stage. The presence of more than one cultivar in the same seed lot results in a loss of varietal purity and compromises the genetic quality of the commercialised product. This is a challenging issue with few solutions available for mitigation on an industrial scale. The cultivar portfolio of the Agronomic Institute of Campinas (IAC) is the most widely used in Brazil for peanut production. Hypothetically, the phenotypic diversity of seeds and seedlings from these cultivars may be associated with a "digital identity" traceable through spectral means, which could enable precise differentiation of genetic material mixtures in the post-harvest phase. The objective is to determine whether emerging and non-invasive technologies can identify peanut seeds and seedlings with diverse genetics based on their multispectral profile. The proposed hypothesis will be tested using basic seeds from ten cultivars in the IAC portfolio: Tatu ST, Caiapó, OL3, OL4, OL5, OL6, OL7, 503, 505, and Sempre Verde. The methodology will begin with the characterisation of seed water content, dry matter mass, germination, and vigor. Next, oil content will be determined using nuclear magnetic resonance. Additionally, multispectral images of the seeds will be captured to assess reflectance, autofluorescence, area, length, thickness, colour parameters, texture, anthocyanins, and chlorophylls. For each cultivar, seedlings will also be produced to investigate photosynthetic efficiency, fluorescence, reflectance, anthocyanins, and chlorophylls. The dataset will be integrated into artificial intelligence models to test the potential for autonomous cultivar identification based on the generated digital patterns. By the end of the project, at least three contributions to Brazil's peanut sector are expected: i) development of digital markers to differentiate varietal characteristics of seeds and seedlings; ii) methods for analysing contamination by seeds from other cultivars as a complement to the official varietal purity analysis; iii) construction of fundamental and multidisciplinary knowledge for managing the genetic quality of peanut seeds produced and commercialised in the State of São Paulo. (AU)