Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Genomic selection via machine learning – a game changer for crop improvement
This thesis aims to understand the principles and methods for GS in crops. It includes the development of new GS models based on machine learning and to test the performance of these models.
Genomic selection (GS) is revolutionising animal and plant breeding: The ability to predict the phenotype from genetic data decreases the cycle length, increases the selection efficiency and thus the genetic gain in a breeding program. GS is now being adopted throughout animal and plant breeding programs worldwide. The increasing use of GS is attributable to accessibility of molecular genetic technologies (high-throughput genotyping platforms) and the improvement of data analysis methods (statistical models). Machine learning tools being developed across disciplines are constantly evolving and offer countless opportunities to improve the predictive ability of GS models. This project seeks to use contemporary machine learning models and develop new approaches to maximize the predictive ability of GS models on the basis of existing genotypic and phenotypic data from major crop species.
Genomic selection (GS) is revolutionising animal and plant breeding: The ability to predict the phenotype from genetic data decreases the cycle length, increases the selection efficiency and thus the genetic gain in a breeding program. GS is now being adopted throughout animal and plant breeding programs worldwide. The increasing use of GS is attributable to accessibility of molecular genetic technologies (high-throughput genotyping platforms) and the improvement of data analysis methods (statistical models). Machine learning tools being developed across disciplines are constantly evolving and offer countless opportunities to improve the predictive ability of GS models. This project seeks to use contemporary machine learning models and develop new approaches to maximize the predictive ability of GS models on the basis of existing genotypic and phenotypic data from major crop species.
- Understanding of the principles and methods for
GS in crops
- Development of new GS models based on machine
learning
- Test the performance of these models using
existing genotypic and phenotypic data from a
contemporary wheat and bean breeding programs
- Understanding of the principles and methods for GS in crops - Development of new GS models based on machine learning - Test the performance of these models using existing genotypic and phenotypic data from a contemporary wheat and bean breeding programs
The Machine Learning & Computational Biology (MLCB) and Molecular Plant Breeding (MPB) labs of ETH Zurich are joining forces to further develop a game changing method for crop improvement. The project will bring together the newest machine learning approaches developed at MLCB with existing GS data and capacities of MPB.
The Machine Learning & Computational Biology (MLCB) and Molecular Plant Breeding (MPB) labs of ETH Zurich are joining forces to further develop a game changing method for crop improvement. The project will bring together the newest machine learning approaches developed at MLCB with existing GS data and capacities of MPB.
The applicant will learn how to use machine learning for GS in some of the most important crop species worldwide.
The applicant will learn how to use machine learning for GS in some of the most important crop species worldwide.
A Master student with strong programming skills in Python and R and a strong interest for statistics and agriculture.
A Master student with strong programming skills in Python and R and a strong interest for statistics and agriculture.
For any questions or details, please contact Prof bruno.studer@usys.ethz.ch or Prof karsten.borgwardt@bsse.ethz.ch
For any questions or details, please contact Prof bruno.studer@usys.ethz.ch or Prof karsten.borgwardt@bsse.ethz.ch