Å·ÃÀAV

Artificial intelligence is transforming chemical research, but many industrial teams struggle to move beyond simple predictive models. Graph Neural Networks (GNNs) represent a major step forward by learning directly from molecular structure, enabling more accurate property prediction, reaction modelling, and synthesis planning.

This advanced course is designed specifically for industrial chemists, process scientists, medicinal chemists, and computational R&D teams who want to understand how modern graph-based AI models can support real-world decision making in synthesis and route design.

Over four intensive sessions, participants will learn how molecules and reactions can be represented as structured graph data and how Graph Neural Networks extract chemically meaningful representations. The course moves beyond theory to focus on practical applications, including molecular property prediction, reaction outcome modelling, yield estimation, and retrosynthetic route evaluation.

Participants will gain insight into how modern Computer-Aided Synthesis Planning (CASP) systems operate, including reaction prediction engines, route scoring models, and search strategies used to generate synthetic pathways. A hands-on mini-CASP workflow will demonstrate how graph models can be integrated into route selection and feasibility assessment.

Crucially, the course addresses the realities of industrial deployment. Topics include data quality challenges, handling proprietary datasets, process-scale constraints, model validation strategies, and the limitations of current AI systems. Participants will learn when graph models provide genuine value and when simpler approaches may be more appropriate.

An elective module allows teams to tailor the final session toward either advanced 3D molecular modelling and materials applications, or AI-driven reaction optimisation and integration with flow chemistry systems.

By the end of the course, participants will be equipped to critically evaluate graph-based AI tools, understand their internal logic, and identify clear pathways for integration into existing R&D pipelines.

This course bridges the gap between cutting-edge AI research and practical industrial chemistry.

We are delighted to offer this online course, which will take place over four sessions on the dates and times outlined below:

PDT: 6.00am-9.00am | CDT: 8.00am-11.00am | EDT: 9.00am-12.00pm | BST: 2.00pm-5.00pm | CEST: 3.00pm-6.00pm

– Tuesday, September 22 – Wednesday, September 23

– Tuesday, September 29 – Wednesday, September 30