PHENO_MaizE (2024-2027)

Project title: High-throughput field phenotyping in temperate maize hybrid breeding: how can phenomics improve speed and accuracy of selection?

Program: Prisma, Science Fund of the Republic of Serbia

Coordinator: Maize Research Institute Zemun Polje

Principal investigator: dr Sofija Božinović

Team members: dr Jovan Pavlov, dr Nikola Grčić, dr Jelena Vančetović, Marko Mladenović, Aleksandar Kovačević

Web site: https://phenomaize.rs/

Evaluating and measuring plants across many environments in plant breeding is essential for genetic gain increase but is also resource demanding and biased. High-throughput field phenotyping (HTFP) allows us to study numerous genotypes in multiple environments and time points. PHENO_MaizE goal is to investigate the use of drone-based HTFP in temperate hybrid maize breeding programs. Different training populations, including inbreds and/or test-crosses, will be used to predict traits of interest in four prediction scenarios that are of great interest to breeders. The project will enable us to evaluate digital traits for their use instead of manual measurements, develop prediction models for the traits of interest using image-derived variables and field images per se, and, finally, select the most informative variables and flights during the season.

Ocena i merenje biljaka u puno gajenih sredina je esencijalno za povećanje genetičke dobiti u oplemenjivanju biljaka, ali takođe može biti subjektivno i zahteva puno resursa. Visoko-informativna fenotipizacija u polju (VIFP) nam omogućava da ispitamo puno genotipova u velikom broju gajenih sredina i vremenskih tačaka u toku sezone. Cilj PHENO_MaizE projekta je da istraži korišćenje VIFP uz pomoć drona u oplemenjivanju kukuruza u umerenom klimatskom području. Različite trening populacije će se koristiti, uključujući inbred linije i/ili test ukrštanja, kako bi se predikovale bitne osobine u četiri predikciona scenarija od velikog interesa za oplemenjivače. Projekat će nam omogućiti da ispitamo korišćenje digitalnih osobina umesto ručno merenih, da razvijemo predikcione modele za bitne osobine korišćenjem digitalnih varijabli kao i snimaka polja per se, i konačno da odaberemo najinformativnije varijable i letove tokom sezone.