Cambridge MedChem Consulting

Category: Drug Discovery

  • AI for Drug Discovery

    The latest meeting in the RSC BMCS Hot Topics series is AI for Drug Discovery.

    Artificial intelligence (AI) is becoming an increasingly important part of the drug discovery toolkit, with applications spanning target discovery, molecular design, protein modelling, and clinical development. As the field advances, AI is not only offering new technical capabilities but also changing how researchers approach the challenges of drug discovery. This meeting will highlight recent progress in AI-driven drug discovery, bringing together perspectives from industry and academia. Talks will explore enabling technologies and practical applications of AI across the drug discovery pipeline.

    There is a fantastic lineup of speakers for real leaders in the field. Registration details are here https://www.rscbmcs.org/events/hottopicsai26/

    Chair: Hannah Fowler, RSC
    12:30Opening Remarks
    12:35Andreas Bender, Khalifa University
    Title TBC
    13:20Keishi Kohara, AstraZeneca
    Embedding AI-Assisted Design at AstraZeneca
    13:50 – 14:05Break
    Chair: Silvia Bonomo, Astex
    14:05Astrid Stroobants, Novartis
    Title TBC
    14:35Fraser Cunningham, Recursion
    Title TBC
    15:05Kenneth Atz, Roche
    RingAnalyzer reveals widespread high-energy ring conformations across structure-based drug design workflows
    15:35 – 15:50Break
    Chair: TBC
    15:50Alex Rich, Inductive Bio
    Dose-Driven Lead Optimization in the AI Era
    16:20Maria Castellanos, OpenADMET, Open Molecular Software Foundation (OMSF)
    Democratizing AI for ADMET: Open Data, Community Models, and Live Benchmarks
    16:50Closing Remarks
    17:00Close of Conference

  • Modelling hERG Channel Liability

    A recent paper in JCIM caught my eye “Modeling hERG Channel Liability: From Structural Insight to Highly Accurate Qualitative and Quantitative Models” DOI . The paper includes a discussion of the binding site on the ion channel commenting on “The highly adaptive nature of the hERG ligand-binding site may poses challenges for structure-based approaches, such as molecular docking” together with a regression model and a classification model. The investigation of the binding site highlights a protonated nitrogen and key aromatic interactions.

    They used a final descriptor set comprised 221 atom types and 40 correction factors, capturing whole-molecule features like molecular flexibility, fraction of sp2-hybridized atoms in a molecule.

    They also provided the curated 8000 compound data set as part of the supplementary information. I had a quick look at the 8000 compound data set.

    I imported the molecules into Vortex and calculated a variety of physicochemical properties and looked up development statues and the number of clinical trials reported (from clinical trials.gov) using a couple of Vortex scripts..

    Looking at calculated physicochemical first, as might be predicted basic molecules tend to be more active at hERG, with acids and Zwitterions much less so. There is also some evidence that the more lipophilic molecules are more active.

    Generating a TSNE plot[ https://en.wikipedia.org/wiki/T-distributed_stochastic_neighbor_embedding] and coloured by hERG class (0 less than 10000 nM, 1 is greater than 10000 nM) used in the publication. Firstly it is clear that hERG activity is widely distributed within the chemical space encompassed in the dataset. It is also a nice way to spot where small changes result in a significant modulation of hERG activity.

    Looking at just the compounds that clinical development has been reported it is clear many have reported hERG activity below 10 uM.

    Whilst most of the compounds are in PubChem, 4772 are also identified in patents, 498 are ligands in the PDB (but not necessarily for hERG).

  • MHRA launches AI sandbox to accelerate medicines development and improve safety

    New AI sandbox will help make medicines safer, speed up development, and reduce reliance on animal testing.

    If you are involved in using AI/ML in drug discovery then this initiative could well be of interest.

    The UK will launch a first-of-its-kind initiative to test how artificial intelligence (AI) can help make medicines safer for patients – as announced by the Science Minister Lord Vallance during London Tech Week today (9 June 2026). 

    The programme will explore how AI can improve the assessment of accuracy and safety, better predict risks, and detect effects that existing approaches may not capture.  

    More details are here https://www.gov.uk/government/news/mhra-launches-ai-sandbox-to-accelerate-medicines-development-and-improve-safety

  • Kiin Bio free offer

    As part of my work I’ll be invited to help startups or review spinouts, this is always really interesting to learn about new science or insights from really smart scientists. However, one of the problems is often navigating through unlabelled presentations or disparate folders on different computers containing excel, word, pdfs etc. Simply putting everything in a data room to let 3rd parties try to navigate is not a viable solution.

    So I’m always interested in potential solutions, Rachel Skyner (who I first met when she worked on Fragalysis) highlighted an interesting looking programme. Kiin are offering elected academic and nonprofit teams get one year of free access to the Kiin Pioneer Programme and access to their drug discovery platform and hands-on support from the science team.

    Research teams are generating more data than ever before, but scientific discovery often stalls at the point of hypothesis-generation and decision-making.

    Promising early findings are often spread across papers, datasets, internal notes, and expert judgment. Priorities can be hard to compare, hypotheses difficult to track, and promising signals slow to translate into action.

    This programme is designed for teams who want to make their discovery process faster, more systematic, transparent, and actionable.

    We are especially interested in teams working on questions such as:

    • Which targets should we prioritise, and why?
    • Which hypotheses are worth testing next?
    • How should we interpret conflicting evidence across datasets?
    • Where are the strongest translational opportunities?
    • How can we make complex scientific decisions easier to track, explain, and revisit?

    No cost, no data transfer, all IP stays with your institution, available to academics and non-profits. Applications close August.