Efrain Torres‐Lomas

The paper presents a low-cost imaging and deep-learning pipeline to analyze root system architecture and soil moisture in grapevine rootstocks, revealing genetic and temporal variation in root traits.

This study evaluates Vitis germplasm diversity to predict adventitious rooting traits, identifying genetic loci and developing genomic prediction models for improved grapevine rootstock breeding.

This study uses topological data analysis and deep learning to show that grapevine leaf blade and vasculature shapes are predictively linked, suggesting a one-to-one relationship between the two tissues.

This study uses QTL mapping and a foundation model for phenotyping to identify genetic loci associated with grapevine cluster compactness and yield traits.

The paper develops a high-resolution morphospace model of grapevine leaves to predict synthetic leaves and improve classification accuracy.