José Luis Crossa

This paper introduces a matrix multiplication framework for genotype-by-environment interaction in genomic prediction, showing improved accuracy over traditional methods.

The paper formulates cultivar recommendation as an axiomatic ranking problem and shows that only the geometric mean satisfies certain minimal criteria for consistency and efficiency.

This study improves genomic selection in hexaploid wheat by incorporating additive and epistatic effects, leading to a 3% increase in prediction accuracy for grain yield.

This study evaluates genomic prediction strategies for carotenoid traits in maize, finding that genome-wide markers outperform gene-specific markers and that GP methods are effective across environments.

This paper enhances multi-year genomic prediction of wheat breeding data by integrating environmental associations (EPA) to improve genotype-by-environment interaction predictions.

This paper introduces E-GP, an enviromic assembly method that improves genomic prediction accuracy for yield plasticity by integrating environmental data and optimizing field trials.

The paper optimizes genomic prediction training sets using enviromic data and genetic algorithms, increasing genetic gains by 142% per dollar invested.

EnvRtype provides a software toolkit for integrating enviromics data into quantitative genomics to improve whole-genome prediction accuracy across different environments.

This study evaluates aerial high-throughput phenotyping for indirect selection of grain yield in wheat breeding programs, finding vegetation indices predictive of yield and more effective than visual selection.

This study evaluates genomic prediction models with genotype by environment interaction for kernel zinc concentration in maize, finding that models incorporating G x E interactions improve prediction accuracy.

The paper develops regularized selection indices to improve breeding value prediction using hyper-spectral image data, showing higher accuracy for grain yield than traditional methods.

This study uses hyperspectral reflectance-derived relationship matrices in genomic selection to predict grain yield in wheat, showing improved accuracy when combined with marker and pedigree data.