Björn Mattsson | Computational Chemistry Consultant

Co-folded GPCR structure with small molecule ligand bound in the orthosteric pocket

Good computational chemistry requires knowing when to trust a model and when not to. I combine modern ML tools with physics-based methods, and rigorous benchmarking, to give you answers you can actually act on.

How I can help

Virtual screening & hit identification

Every target is different, so I start by evaluating which computational approach (ligand-based ML, co-folding, physics-based methods, or a combination) is most likely to work for yours. The goal is a ranked shortlist of hits you can take into the lab with confidence.

ML model development

I can help you scope, build and validate machine learning platforms to help answer recurring questions where you are starting to gather data, such as “does the compound bind”, or “does it permeate membranes”.

How I can help

Virtual screening & hit identification

ML model development

Would you like to get in touch?