Abstract

autoTRAIT will establish a generative AI–synthetic biology pipeline for plant gene regulation, for optimization of crop performance. By combining foundation models, supercomputing, and Plant STARR-seq, the project directly contributes to the BioSC focus area “Smart management of plant performance.” The project unites expertise in AI and bioinformatics with synthetic biology and experimental validation. Together, the partners will: (a) Fine-tune genomic foundation models on plant-specific data; (b) Interpret regulatory logic with explainable AI; (c) Generate tunable gene regulatory segments relevant for flowering time, plant height, and drought response; (d) and validate large-scale libraries of such segments of Arabidopsis, maize, and Nicotiana experimentally. The outcome will be a self-optimizing Design–Build–Test–Learn pipeline for regulatory DNA engineering. The project contributes to laying the foundations for long-term AI applications in climate-friendly agriculture.

SEED FUND 3.0 Project coordinator

Dr. Jędrzej Jakub Szymański
IBG-4 Omics-/Data-based Bioinformatics
Forschungszentrum Jülich
email: j.szymanski[at]fz-juelich.de

Partners

Dr. Jędrzej Jakub Szymański, Prof. Dr. Björn Usadel, IBG-4 Omics-/Data-based Bioinformatics, Forschungszentrum Jülich
Dr. Tobias Jores, Prof. Dr. Matias Zurbriggen, Institute of Synthetic Biology, Heinrich Heine University Düsseldorf

Funding period

01.01.2026 - 31.12.2026

Funding

autoTRAIT is part of the NRW-Strategieprojekt BioSC and thus funded by the Ministry of Culture and Science of the German State of North Rhine-Westphalia.