Together with academic and industry partners, Computomics will help to access native diversity in landraces to improve quantitative traits relevant for crop production in maize.
After the highly successful completion of MAZE Phase 1, where a team of researchers gained deep knowledge on doubled-haploid flint lines, Phase 2 officially started in February 2020 and will run for 3 years.
The overall goal is to include the native diversity and stress tolerance present in landraces to secure sustainable crop production. During Phase 1, doubled-haploid lines were generated and evaluated under different environmental stress conditions. Computomics will support the project with structural and functional annotations. Functional annotations will be performed with our automated, scalable pipeline AnnoScore which uses artificial intelligence technologies. With our skills in high-quality annotations, candidate genes and alleles important for environmental stress responses can be detected and explored with respect to their potential to improve elite germplasm.
Within the scope of MAZE Phase 2, prediction models based on artificial neural networks (ANN) will be optimized, to extract GxG, GxE and GxGxE interactions. Computomics will contribute with expertise in machine learning-based performance prediction and the extraction of higher-order features that are important for plant performance.
The results of this project will be freely available to researchers and breeders, to include the gained knowledge to generate optimized germplasm and breeding schemes. Computomics is excited to contribute to this project securing crop production in fast changing environments.
The project is led by Prof. Dr. Chris-Carolin Schön from Technische Universität München.