AIchemy’s Monthly Webinar Series – January 2026

We are delighted to welcome you to our AIchemy Hub’s monthly webinar series.

This month’s talks:

Andrea Dimitracopoulos – COO & Co-founder, Deepmirror

Talk Title: Tackling Drug Design Challenges with Molecular Assay Fingerprints

In the journey from idea to clinic, drug discovery teams must balance potency with a wide set of ADMET constraints (solubility, permeability, CNS penetration, clearance, toxicity risk, and more). AI can reduce the cost and time of small-molecule drug design by improving both property prediction and molecule prioritisation across the design–make–test–analyse (DMTA) cycle. In this talk, we introduce molecular assay fingerprints as a learned representation for modelling assay endpoints. This approach combines pre-training on a large assay database with programme-specific fine-tuning to improve prediction accuracy. It also leverages Gaussian processes to quantify predictive uncertainty, which can then be used to prioritise compounds and explore chemical space efficiently – accelerating programme progression. Benchmarking on private drug programmes illustrates how assay fingerprints can improve predictive accuracy for both potency and key ADMET properties, dramatically reducing the number of design cycles.

Federico Ottomano – Imperial College London

Talk Title: Scalable molecular elucidation from IR and NMR spectroscopy using Machine Learning