IMAGE-BASED PHENOTYPING APPLIED TO GRAPEVINE BREEDING

• Grant number: 23/09468-7
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: December 01, 2023
• End date: November 30, 2025
• Field of knowledge: Agronomical Sciences - Agronomy - Crop Science
• Principal Investigator: Mara Fernandes Moura Furlan
• Grantee: Felipe Roberto Francisco
• Host Institution: Instituto Agronômico (IAC). Agência Paulista de Tecnologia dos Agronegócios (APTA). Secretaria de Agricultura e Abastecimento (São Paulo - Estado). Campinas , SP, Brazil
• Associated scholarship(s): 24/19464-1 - Phenological characterization of grapevines through computer vision, BE.EP.PD

Abstract

The grapevine culture has been gaining prominence in Brazilian fruit growing, passing, in recent years, from an exclusive cultivation in temperate zones to a great alternative also in subtropical and tropical regions. Therefore, it is necessary to implement biotechnological tools in plant breeding programs to quickly make available to farmers new cultivars with aptitude for processing where the final objective is the juice. The phenotyping of traits in fruit breeding programs is performed manually, being an expensive and time-consuming activity, which demands intensive labor and requires extensive experience from the evaluator. In order to improve the efficiency of the phenotyping of the traits of interest, methodologies based on image analysis and processing were developed. These methodologies enable non-destructive measurements, with little manpower, savings in financial resources, speed and precision. The present proposal aims at the application of image-based phenotyping in the genetic improvement program of Instituto Agronômico (IAC) in Genomic Wide Association Study (GWAS) and Genomic Selection (GS) techniques). The GWAS and GS techniques use genomic marker data to aid in the identification of superior grapevine genes and genotypes, respectively. In order to explore the relevance of these techniques, we will use 288 accessions belonging to the Germplasm Bank of the IAC. In the present work, the following phenotyping methodologies will be evaluated: 1) Use of unmanned aerial vehicles (UAVs) equipped with RGB and multispectral sensors for high-throughput phenotyping of grapevine plants; 2) Image-based phenotyping of agronomic traits under field conditions; 3) Phenotyping based on images of vine clusters and berries; 3) Prediction of sugar content, pH and anthocyanin in grape must by digital images; 4)Use of digital images for recognition and prediction of disease incidence in grapevines (Plasmopara viticola, Erysiphe necator and Elsinoë ampelina). The potential of GS and GWAs techniques will be evaluated in explaining heritability, detecting significant associations and predicting phenotypes of agronomic traits. The expectations of the implementation of the GS-GWAS techniques assisted by images applied to the genetic improvement of the vine are great, because it opens a concrete perspective of carrying out direct selection with the possibility of increasing the genetic gain per unit of time, and shortening the time necessary for selection of superior genotypes