
Comp. matsci postdoc at @UCLouvain_be & @Matgenix interested in materials discovery and data management for open science, also https://teapaw.org/t.co/sDCho3BPom
King's Lynn, UK / Brussels, BE
Joined May 2011
- Tweets2.5K
- Following1K
- Followers349
- Likes1.8K
In sunny Boston for #f24mrs! Just putting the "finishing" touches on my talk from the public library ready for Tuesday at 13:30. Come listen to me talk about OPTIMADE, datalab and their relevance to broadening access to AI in materials research!
2
13
620
For those of you looking for a big helping of LLMs for Thanksgiving, here you go! 🤖🦃 (link in next post)
In this work, we release 34 examples (each with open code!) demonstrating applications of LLMs in materials science and chemistry across:
🔶molecular and material property prediction;
🔷molecular and material design;
🔶automation and novel software interfaces;
🔷enhanced scientific communication and education; 🔶research data management and automation;
🔷hypothesis generation and evaluation; and
🔶knowledge extraction and reasoning from scientific literature
2
102
Matthew Evans / @[email protected] retweeted
We have been on Bluesky for a while and we would love for you to join us there as well! 🦋
Follow us ▶️go.bsky.app/QUQJbsd
Matthew Evans / @[email protected] retweeted
Presented #semanticClimate's work in 5-starification of climate documents for its sustainable future. My takeaway from the panel: #TheTechWeWant is simple, plural and fosters community. Software should be rooted around the planet with goals other than economic growth. @OKFN
Our paper on the latest OPTIMADE developments is now in print in @digital_rsc (doi.org/10.1039/D4DD00039K), accompanying the latest version of the specification (changelog: github.com/Materials-Consort…) [1/2]
The large-scale use and exchange of data is a prerequisite for AI in materials research. A major international collaboration, including #CECAM and @nccr_marvel, now presents an extended version of the #OPTIMADE standard.
Read more ➡️ cecam.org/news-details/inter…
As always, thanks to all the other authors and contributors who have volunteered their time over the past years. Looking forward to seeing how people build on our efforts! [3/2]
1
3
115
If you are interested in getting involved we still hold open monthly meetings online (details at optimade.org), and fingers crossed there will be another CECAM workshop next year
teapaw.org/cecamEvents/status/180…
The workshop "Open Databases Integration for Materials Design" returns to CECAM-HQ (& online) this week!
The @optimade_ consortium was first established to create a comprehensive #API capable of accessing all #materials #databases, to address the issue of scattered databases.
1
2
168
We (@jd_boc+Ben+Vraj) did another LLM hackathon on datalab, here is our demo showing an autonomous agent using our Python API to perform some useful tasks in experimental data management:
youtu.be/mbeKdJJmtWA
Links:
- github.com/bocarsly-group/ll…
- github.com/the-grey-group/da…
thismaterialdoesnotexist.com -- or does it? Potentially fun+cool way to explore the latest materials discovery datasets. If there's enough interest I can do something with the statistics (currently serving random structures from a subset of Deepmind's Gnome)
Dataset (for now): github.com/google-deepmind/m… and github.com/ml-evs/optimade-g… Code: github.com/ml-evs/this-mater… mostly using github.com/materialsproject/…
Matthew Evans / @[email protected] retweeted
Check out our latest work relying on Machine Learning to accelerate the discovery of electrides #materialsscience #mlchem #compchem pubs.acs.org/doi/10.1021/jac…
This is now running at optimade-gnome.odbx.science/…, still a few wrinkles to iron out but you can easily query the dataset programmatically with the OPTIMADE filter language, might be useful for the ongoing reproduction efforts...
Replying to @funroll_loops
I hope we can make this dataset accessible via OPTIMADE (optimade.org) for wider and more accessible dissemination!
For example, here's the output for the newly discovered CO₁₄ phase... (think I'm going to have to bring this down and just use the 160k SCAN dataset tomorrow...)
1
262
Crap, I bought it... if anyone wants to a hack on this project over the weekend, I'm aiming to set this up to collect stats on citizen's opinions of which phases should be prioritized for further study
To clarify, this is a bit tongue-in-cheek and is inspired by thispersondoesnotexist.com -- I think there are some very clearly wacky phases amongst the good stuff in this dataset and it would be nice to quantify/communalize that
2
233



















