Skip to content

Cambridge Cheminformatics Newsletter, April/May 2023

Dear All,

I would like to circulate some current Cheminformatics- (and related) news to everyone as follows. If you have information from your side for distribution please just let me know, and I am happy to include it on the next occasion!

So here we go…


7 June 2023
Cambridge Cheminformatics Meeting
Cambridge, UK and Online (Hybrid Mode)

In person at the Cambridge Crystallographic Data Centre on Union Road
Registration (for Zoom attendance):


Structure-based Drug Design with Equivariant Diffusion Models
Charlie Harris, University of Cambridge

DECIMER: Deep Learning for Scraping, Curating and Registering Compounds From the Primary Literature
Kohulan Rajan, Jena University

Distributed HPC Workflows with Covalent
Will Cunningham, Agnostiq

15 June 2023
Berlin Modeller Stammtisch
Berlin, Germany

19 June 2023
Cambridge Alliance on Medicines Safety (CAMS) Event: Translation of Safety in the Clinic
Virtual Event

26/27 June 2023
Industry Symposium on “AI in the Life Sciences”
Bonn, Germany and Remote

4/5 September 2023
6th Artificial Intelligence in Chemistry Symposium
Cambridge, UK

6 September 2023 – hold the date!
Cambridge Cheminformatics Meeting – Stay in Town after the AI in Chemistry Symposium!

19-22 September 2023
Summer School: Artificial Intelligence for Medicinal Chemistry and Drug Design
Muenster, Germany

Job Seekers

Laboratory digitalisation expert with sound knowledge in topics like FAIR Data management, Digital transformation and driving data driven decisions and culture. Main skills include project management, people management, software implementation life cycle, and laboratory automation. Notable knowledge of Pharma, Biotech and Drug discovery processes.

Please contact me and I will bring you in touch (


Tenure track position on AI and multiscale modeling in biology and bio-medicine
University of Montpellier
Montpellier, France

Cheminformatics Senior Scientist, ADME
San Francisco Bay Area or Remote

Richter Gedeon
Budapest, Hungary

Principal Scientist Computational Chemistry
Cancer Research UK (CRUK)
Glasgow, Scotland

Computational Chemist
Cambridge, UK

Director Computational Molecular Design; Machine Learning Scientists
Berlin, Germany; Cambridge/MA

IT Early Solutions Scientific Compute Lead
Braine-l’Alleud, Belgium

Head of Data Science
Wild Biotech
Tel Aviv, Israel

Software Engineer (Machine Learning)
Isomorphic Labs
London, UK

Senior Research Scientist I/II, Cheminformatics (Augmented Molecular Design)
Chicago, IL

Postdocs – Geometric Deep Learning and AI/ML for Image Analysis
Madrid, Spain and Beerse, Belgium

Internships Computational Toxicology  
Basel, Switzerland;-from-May-2023-for-9-12-Months


Spring 2023 UK-QSAR Newsletter

News from in and around the field

chemfp 4.1 is out
“chemfp is a set of command-line tools and a Python package for working with binary cheminformatics fingerprints, typically between several hundred and a few thousand bits long.”

Breeze 2.0: an interactive web-tool for visual analysis and comparison of drug response data

So is it a hit, or not?

Benchmarking Refined and Unrefined AlphaFold2 Structures for Hit Discovery
“[…] we find that unrefined AF2 structures deliver similar enrichments to that of an apo experimentally derived structure, significantly below the enrichments using an experimentally derived holo structure. Meanwhile, the application of IFD-MD [an induced-fit protocol] can induce a binding site conformation that delivers enrichments much closer to the holo structure.”

Are Deep Learning Structural Models Sufficiently Accurate for Virtual Screening? Application of Docking Algorithms to AlphaFold2 Predicted Structures
“…we strongly recommend to include some post-processing modeling to drive the binding site into a more realistic holo model” (always good if studies agree!)

Infinite Physical Monkey: Do Deep Learning Methods Really Perform Better in Conformation Generation?
Sample sizes matter when evaluating docking and scoring

ChemCrow: Augmenting large-language models with chemistry tools
Interfacing domain knowledge with Large Language Models

“BigSolDB contains 54273 individual solubility values, 830 unique molecules and 139 individual solvents for the temperature range from 243.15 to 403.15K and at atmospheric pressure”
Thanks for making this information available – ideally pH would be added in future versions of the database

“Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl”
As the title says (but be aware of biases of your chemical datasets!)

Using Transcriptomics and Cell Morphology Data in Drug Discovery: The Long Road to Practice
Might be useful to those interested in systems-based readouts and their use for drug discovery

… beyond Cheminformatics …

Catalyzing next-generation Artificial Intelligence through NeuroAI
The next big thing to come?

Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training
Adam is outdated… it’s time for Sophia!

The fallacy of the average: on the ubiquity, utility and continuing novelty of Jensen’s inequality
“On average two people walk over a bridge… but actually one is flying, and one is drowning” (one of my favourite sayings)

60 Years of Artificial Intelligence at Stanford
As the title says  

OpenAI Knows GPT-4 Is Dangerous—but Won’t Do a Damn Thing About It
This reads quite ‘old’ in a way already despite being just two months old, but it doesn’t mean it’s entirely outdated

Denied by AI: How Medicare Advantage plans use algorithms to cut off care for seniors in need
Data and algorithms are one thing, validation and deployment another

Humility in Design May Be Hubris in Science: Reflections on the Problem of Slodderwetenschap (Sloppy Science)
Science isn’t easy if you really think about it

Laws of Tech: Commoditize Your Complement
What Microsoft did with IBM, and why (some) commercial companies are only too happy to contribute to open source projects

… and really beyond Cheminformatics

“The expert who pioneered ‘quantitative easing’ has seen enough: Central banks are too powerful and they’re to blame for inflation”
And yes, inflation we have

Language Model Validation Special Section:

GPT-4 absolutely crushes Bard in a rap battle
I would agree with that judgement!

And it is even learning to…
… draw unicorns in TikZ 🙂

I believe this is all from my side for now – if you have any information for me to circulate, or wish to present at one of our next Cambridge Cheminformatics or Digital Science for Drug Discovery Meetings, please just let me know, cheers!

Best wishes,


Leave a Reply

Your email address will not be published.