Postdoctoral Position in AI-Driven Drug Design

Developing and adapting machine-learning approaches for structure-based and generative molecular design. Integrating physicochemical information, including protein–ligand interaction features, into generative AI workflows. Developing computational workflows for closed-loop DMTA cycles in which experimental affinity, selectivity, and molecular-property data are continuously used to improve the next generation of proposed molecules. Applying and validating the developed approaches prospectively in the design and optimization of serine protease inhibitors. Collaborating closely with computational scientists, chemists, and biologists within the international project consortium. Contributing to scientific publications, presentations, and project reporting.

A fully funded Postdoctoral position is available in the Computational Pharmacy group at the University of Basel within an international Innosuisse research project on AI-driven closed-loop drug discovery. The project aims to establish an integrated Design–Make–Test–Analyze (DMTA) platform combining generative AI, ultra-large synthetically accessible chemical spaces, physics-informed molecular representations, off-target prediction, and experimental feedback. The developed methods will be applied in iterative prospective drug-discovery cycles, with a serine protease from the complement system serving as a real-world lead-optimization case study. The successful candidate will play a central role in the computational and AI components of the project and work closely with our international and industrial project partners.

• PhD in Computational Chemistry, Cheminformatics, Computer Science, Physics, or a related discipline.

• Strong background in machine learning and deep learning.

• Strong programming skills, particularly in Python.

• Experience in at least one of the following areas: molecular generative AI, cheminformatics and molecular representations, structure-based drug design and protein–ligand modeling.

• Experience with molecular modeling and a good understanding of the physicochemical principles governing molecular recognition is highly desirable.

• A strong publication record in internationally recognized, high-quality venues is required, such as leading journals in computational chemistry (e.g., JCTC, Journal of Chemical Physics) or top-tier machine-learning conferences (e.g., ICLR, ICML, NeurIPS), as appropriate to the candidate's research background.

• Fluent verbal and written communication skills in English.

• Highly motivated, independent, and collaborative researcher with an interest in working at the interface between methodological development and prospective drug discovery.