Profile avatar
jpoverington.bsky.social
Work at drughunter.com - Drug Discovery, Data Science, Cheminformatics, Computational Medicinal Chemistry, Drug Repositioning, Structural Bioinformatics, CADD, SaaS, DaaS, AI/ML https://scholar.google.com/citations?user=52VMFeYAAAAJ
36 posts 764 followers 1,255 following
Prolific Poster

Interested in high-throughput chemistry, structural biology and cell biology to understand biological mechanisms? We have a PhD position available with @baylisslab and @MWrightLab ⁦https://phd.leeds.ac.uk/project/2067-high-throughput-discovery-of-lysine-directed-probes-for-protein-kinases⁩

New work laptop. Don’t you love it when you have to adjust to a new keyboard and all that muscle memory is worthless.

I’ll shortly be recruiting a few roles for the new team. Data science, cheminf, python, NLP, NER, dictionaries, ETL, that sort of stuff. Remote only. Feel free to reach out informally if you’re interested.

So, I start a new job Monday. Thought I’d spruce up the LinkedIn profile and make me look more professional (ha ha). So I tried an AI body shot app - must admit I do now look fancy. Imagine the disappointment though when I rock up for a meeting now.

The AI search agents out there are really bad. Here's an example from this morning that looked odd, so I checked. Turns out it's simply wrong. Do the search again on same computer and it's yet a different answer. Search engines should search, not guess in my view.

Some professional news, I’m shortly starting a new position at drughunter.com. Really looking forward to working with team their and helping their mission!

I've done my Venn - thx @winhide.bsky.social for the Venny recommendation. This is an overlap presence of chemical structure in 3 'gold standard' drug databases for approved protein kinase small mol drugs. There's 1 approved drug for which the structure is in none of them.

The internet, or more specifically search engines, are now a dumpster fire - I wanted a simple Venn diagram maker, just paste in some lists, plot of out. Simples. Haven't found a single one yet that doesn't need signing in, a trial account set up or some nonsense. Any recommendations?

Missed this paper last, turns out that scientific publishing and 'open access' is pretty lucrative. Who'd have guessed. Publishing science really needs to change. arxiv.org/abs/2407.16551

A different plot on the same dose/potency data as previously - well an expanded version of it. This time dose rank (in moles) vs IC50 potency rank.

Dose prediction for drugs is important and challenging - you'd think potency would dominate - more potent -> lower dose. It's not that simple - here's a plot of required daily dose (in mM) against median target potency (in nM). This plot is for orally-dosed small-molecule protein kinase inhibitors.

"Artificial Intelligence in Retrosynthesis Prediction and its Applications in Medicinal Chemistry" in JMC pubs.acs.org/doi/10.1021/...

Just starting on reading Hallelujah Moments by Cordes, following a recommendation from Martin Shkreli. Great so far. a.co/d/fDvFSpX

To fans of British Alternative musics from the 70s to 80s… today is the Anniversary of the Movement of 24th January. I was 13 years old, an impressionable teenager.

I'm a bit of a drug naming nerd, the system used in China has always been opaque to me; here's a great overview of some of the complexities of generic/statutory drug naming in China. 'Anticompetitive effect of drug name trademark registration: lessons from China' www.nature.com/articles/s41...

Site-saturation mutagenesis of 500 human protein domains www.nature.com/articles/s41...

Selective targeting of Cys using protein structure prediction and ML pubs.acs.org/doi/abs/10.1...

Accurate predictions on small data with a tabular foundation model www.nature.com/articles/s41...

Ligand discovery by large scale chemoproteomics www.nature.com/articles/s41... #biosky

Interested in a PhD combining high-throughput synthesis, structural and cell biology to understand kinases? Deadline to apply for a project with Richard Bayliss, Helen Matthews and me: January 6th! www.findaphd.com/phds/project...

Feeling festive with @thamesvalleyrsc.bsky.social . What's on your Christmas menu? 🦃🎄 #adventcalendar @compoundchem.com #ChemSky #EduSky

This week's paper (well, last week's paper, but it took time to process) returns to the theme of reference protein coding genes. We wanted to know if the human proteome was more streamlined after we merged RefSeq, @gencodegenes.bsky.social and UniProtKB back in 2018. www.biorxiv.org/content/10.1...

And another in the occasional series “American’s can pun”. This I found so funny. Jurassic Bark!

I thought it was going to be a boring Xmas party. But it’s only at the Houston Natural History Museum!

Now at the dog pub. I don’t know who is more excited, Tiffin or me.

And I thought the Brits were kings of shop name puns.

I left work this Tuesday. Do you know what I’m missing most? That’s right my .vimrc file. #ViGOAT

Happy to announce that I’ll be speaking at the CSSB symposium in Hamburg, early May. Looks like there’s a great lineup of speakers, and then me! Conference details at symposium2025.cssb-Hamburg.de/

There's a new #RDKit blog post introducing some new functionality that I'm really excited about: doing efficient substructure and similarity searches in very large chemical libraries: greglandrum.github.io/rdkit-blog/p... #ChemSky

The Lego Star Wars minifigure curation is ongoing, about two thirds done, getting trickier matching bodies and legs now. There must be a nerdy online resource for this? #LegoStarWars

My guess is that this post is a joke to 50% of scientists and serious to the other 50%.

The results of #CASP16 are online, pdf files of the presentations will be made available next Wednesday through the CASP website. The release of talk recordings will follow soon afterwards. Big thanks to all participants for an inspiring conference! 🧪 🧬 predictioncenter.org/casp16/

Forgot to stick the #ChemAdvent24 tag on this one yesterday, and a couple of hydrogens had also gone walkies (thanks for spotting, @stareatair.bsky.social!), so here's a corrected version!

My OCD and need to curate and organise the world around me is in full flow. Here are some of my ‘sons’s’ Lego Star Wars mini figures (given his age, not sure I can use this displacement strategy!)

Thinking of getting some reference manager software, I have a very large pdf collection though - 100k+. Will Papers cope? Any recommendations?

Meant to post a picture of this years outside lights. Boy was it a lot of work to set them up, on ladders (for the guys that installed them 🤣)

greglandrum.github.io/rdkit-blog/p...

#compchem Good read: Rapid prediction of molecular crystalstructures using simple topological andphysical descriptors #compchemsky www.nature.com/articles/s41...

This weekend the xmas trees went up. As per tradition, we had the annual argument over how long per day the lights should be on.

Bought a cacao pod today. Surprisingly large, but the most incredible thing to eat. Lovely sweet slimy goo, then a soft seed with the most in two or top chocolate notes. Off to buy some more to ferment….

www.nature.com/articles/s41...

Thanks for everything Professor Bickerton! 👏 🥹🧡 www.linkedin.com/posts/richar...

I just joined Bluesky as well... let's rebuild the same nice community in the cheminformatics/comp chem/bioinformatics/drug discovery etc. areas that 'the other place' was in the good old days 😊! Looking forward to reconnecting with people here, Andreas