Andrew White, Edison Scientific

Jackson Burns, MIT
CheMeleon: Deep Learning Foundation Models from Classical Molecular Descriptors

Layla Hosseini-Gerami, Ignota Labs
Towards improved activity cliff prediction in homologous chemical series

Patrick Walters, OpenADMET
OpenADMET – Elevating ADMET Prediction Through Open Science

Gabriele Corso, Boltz
Boltz: Towards Accurate Biomolecular Modeling and Design

Lauren Delong, University of Edinburgh and Enveda
Neurosymbolic AI for Drug Discovery: Current State, Limitations, and the Path Forward

Teresa Head-Gordon, University of California, Berkeley
Machine Learning Foundations and Large Language Models are Here for Molecular Discovery

Jacqui Cole, University of Cambridge
Data-driven Materials Chemistry

Kathryn Giblin, AstraZeneca
From Generative AI Methods to Delivered Chemical Series: Experiences of Applying AI to Real Drug Projects

Julia Westermayr, University of Leipzig

Giulio Volpin, Bayer
Data Science-Assisted Workflows for Reaction Optimization in Process Chemistry

Rebecca Paul, Isomorphic Labs

Registration is open here

https://www.eventsforce.net/hg3/351/register

There is also exhibitor space available.

https://www.eventsforce.net/hg3/351/register

Exhibition stand package is priced at £1,400 and includes:
– A six-foot trestle table and chair(s);
– Access to electricity and Wi-fi;
– Logo inclusion in pdf delegate handbook and rolling slides;
– Exhibitors promotional page in the pdf delegate handbook;
– Logo included in the communication emails to delegates;
– One exhibitor stand staff with access to the technical sessions and Conference dinner (excluding accommodation)

https://priorlabs.ai/technical-reports/tabpfn-3

TabPFN is a foundation model  trained on around 130,000,000 synthetically generated datasets that mimic “real world” tabular data. These datasets sampled dataset size and number of features, both classification and regression tasks, and Gaussian noise was added to mimic real-world complexities. This can then be used to build models for small- to medium-sized datasets with up to 10,000 samples and 500 features and is claimed to be superior to other methods.

A new performance standard. On the standard tabular benchmark TabArena, a forward pass of TabPFN-3 outperforms all other models, including tuned and ensembled baselines, by a significant margin, and pareto-dominates the speed/performance frontier. TabPFN-3 also scales to more diverse datasets: it ranks first on datasets with many classes, and beats 8-hour-tuned gradient-boosted-tree baselines on datasets up to 1M training rows and 200 features.

A fantastic step-by-step breakdown of an example Python pipeline used to build a ligand-based 3D pharmacophore model using free, open-source tools.

The Cambridge Cheminformatics Network Meetings are free to attend and open to all. We start our meetings in the afternoon at 4pm (UK time) with a series of short scientific talks, currently in hybrid mode on Zoom and at the Cambridge Crystallographic Data Centre (CCDC) in person,  usually followed by a social at the Panton Arms and an ‘Online Worldwide Cheminformatics Pub Night’!

Details are on the website https://www.c-inf.net

All the data is freely available here https://openbind.uk/documents-and-tools/

The UK‑led OpenBind initiative has reached a major milestone with the announcement of the release of its first publicly available dataset and predictive AI model, a groundbreaking step toward accelerating the discovery of new medicines using artificial intelligence.

https://registrations.hg3conferences.co.uk/hg3/frontend/reg/thome.csp?pageID=148219&ef_sel_menu=2770&eventID=363

Quantum computing has the potential to revolutionise computational science but has yet to do so. What has it done for chemistry already, what can it do now, and what might it be able to do in the future? This meeting will bring together people working on the applications of quantum computing to chemistry. It will focus on current capabilities and analyse the road to utility. Is quantum computing useful now in addressing specific molecular problems? When will we reach the milestones needed to expand its utility? It will be useful for chemists working in electronic structure theory, molecular design, property prediction and modelling chemical processes. It will answer the questions of what can be done now in each of these fields, and how long we might need to wait for practical quantum computing tools. 

This meeting is now accepting oral and poster abstract submissions.

The core bet is simple: biological data is expensive. Text and image models often improve by consuming more data, but protein structure data is far more constrained, far harder to generate, and far more sensitive to leakage. If we want better biological foundation models, we need architectures and training methods that make stronger use of the data we already have.

Full details are on GitHub https://github.com/ChrisHayduk/nanoFold-Competition

PyChem is a desktop chemistry application and Python library that combines molecular visualization (like PyMOL) with cheminformatics primitives (like RDKit). Unlike most tools in the space, PyChem is pure Python with NumPy. There is no C++ extension, no RDKit dependency, no OpenBabel binding. Every feature — SMILES parsing, 3D coordinate generation, force field optimization, descriptor calculation, Shrake-Rupley SASA, ring perception, protein cartoon rendering — is implemented from scratch and readable end-to-end.

Sweet J can be downloaded here https://www.inmr.net/sweetj.html

Giuseppe Balacco
A Desktop Calculator for the Karplus Equation
J. Chem. Inf. Comput. Sci., 1996, 36 (4), pp 885–887
DOI: 10.1021/ci950227r

Details are here https://nebius.com/ai-discovery-award Submission deadline is April 30 2026.

1st place $100,000, 2nd place $50,000 and 3rd place $30,000.

Each participant should have a legal entity, a functioning website, product MVP and GTM strategy.