Senior ML Scientist @ BioPhoundry | Machine Learning | Protein Design
Edinburgh, Scotland
Joined November 2020
- Tweets78
- Following549
- Followers281
- Likes358
Michael Stam retweeted
All designs were scored with MD magic from @EvaNotari and our very own "Stam score" (@mjstam ), and we filtered out any candidates that showed non-specific co-folding.
Fingers crossed for the lab experiments!
Michael Stam retweeted
We (@ChrisWellsWood Wood's lab) have taken the lead at the @adaptyvbio competition against the Nipah Virus!
Give us your ⭐️ for our other designs here:
proteinbase.com/collections/…
Michael Stam retweeted
The protein design competition community vote is now open!
This edition, we received over 11,000 proteins from over 600 designers. That's more than 5x increase from our last competition. In other words, there's a lot to explore!
How to vote:
➡️ Head to proteinbase.com
🧐 Browse the submissions
⭐ Star your favorites
You can vote for as many as you'd like.
The top submissions (up to 200 proteins in total) based on community votes will be experimentally tested in the lab.
Voting closes December 10th 11:59 PM PST .
Michael Stam retweeted
Introducing BenchBB and the community paper of the Protein Design Competition 🎉
We wrote a community paper about our Protein Design Competition, teaming up with your favourite protein designers from both rounds. Excited to announce that this is now on Biorxiv 🙌
Michael Stam retweeted
The Protein Design Archive (PDA): insights from 40 years of protein design go.nature.com/4iNRFLI
Michael Stam retweeted
Our paper on computational design of chemically induced protein interactions is out in @Nature. Big thanks to all co-authors, especially Anthony Marchand, @St_Buckley and @befcorreia
nature.com/articles/s41586-0…
Michael Stam retweeted
A framework for evaluating how well generative models of protein structure match the distribution of natural structures.
@PossuHuangLab
Michael Stam retweeted
🎇 We're kicking 2025 off with a Protein Design Archive update, adding structures such as miniaturized electron-transfer proteins, small molecule and protein binders, beta barrels and hairpins, oligomers, and more! Check out all of them at pragmaticproteindesign.bio.e… @ChrisWellsWood
🙌🧬 Incredibly proud of the Wells Wood lab's results in round 2 of the @adaptyvbio protein design competition! All of our designs expressed and one of them bound to EGFR.😀🎉 Check out the list below for the different models and software that we used to do this. 👇 (1/4)
PROTEIN DESIGN COMPETITION ROUND 2
The long awaited results are here -- and they don’t disappoint!
📈 We doubled the number of proteins we tested in our lab from 200 to 400!
🧬 Out of those 400 proteins, 378 expressed (95% expression rate!)
🚀 Out of those 378 expressed proteins, 53 did successfully bind the target protein EGFR (that’s a 14% success rate, more than 5x of the success rate of the previous round just 2 months ago!)
💪 The best binders reached single-digit nanomolar affinities which is within the range of the commercially sold Cetuximab antibody by Merck
😎 30 out of 130 protein designers managed to design at least one binding protein
More details ⬇️
Additional tools:
NetSolP - Protein sequence solubility prediction doi.org/10.1093/bioinformati…
ColabFold - Accessible protein folding doi.org/10.1038/s41592-022-0… (3/4)
Thanks to everyone from the Wells Wood lab who got involved in this submission. 🚀🧑🔬 @ChrisWellsWood, @leocastorina, @MartaChronowska, @s_mert_unal, Tadas Kluonis, Zhaocheng Liu, Julija Sabulyte, Eugene Shrimpton-Phoenix. Can't wait for the next design round @adaptyvbio. 😉 (4/4)
Michael Stam retweeted
PROTEIN DESIGN COMPETITION ROUND 2
The long awaited results are here -- and they don’t disappoint!
📈 We doubled the number of proteins we tested in our lab from 200 to 400!
🧬 Out of those 400 proteins, 378 expressed (95% expression rate!)
🚀 Out of those 378 expressed proteins, 53 did successfully bind the target protein EGFR (that’s a 14% success rate, more than 5x of the success rate of the previous round just 2 months ago!)
💪 The best binders reached single-digit nanomolar affinities which is within the range of the commercially sold Cetuximab antibody by Merck
😎 30 out of 130 protein designers managed to design at least one binding protein
More details ⬇️
Michael Stam retweeted
🚀 Protein Design Competition: Design phase completed -> Wet lab validation started
We’ve received well over 1000 protein designs by 130 amazing proteins designers for the 2nd round of our competition.
With so many people enthusiastic about designing novel proteins and so many interesting submissions, we decided to double the number of proteins that we will test in the lab!
So instead of just 200 proteins we will experimentally characterize 400 novel proteins in our lab and open-source all the results.
Here’s how we selected them:
- Top 100 designs according to the computational selection metric
- For the other 300, we picked between 1 and 5 designs per protein designer, depending on the quality and novelty of their design process descriptions
That means that every participant that wrote at least some description of their design process will have at least one of their proteins validated in the lab! 🎉
Excited to announce that our team's designs ranked among the top 30-57 positions in the @adaptyvbio protein design competition, distinguishing themselves among over 1,000 submissions! So proud of our team! 😀
WE DID IT! All our designs for the @adaptyvbio competition using TIMED in all of its flavours (new and old) were in the top 30 - 57 out of 1000+ designs!
Congrats to the Wells Wood Lab: @ChrisWellsWood Eugene, Julija,@mjstam @MartaChronowska, Mert, Tadas, Zhaocheng!
Michael Stam retweeted
There is a new tool on BioIcons to turn any protein (in PDB format) into a 2D vector illustration in SVG format. You can upload any PDB file or import from PDB, AlphaFoldDB or ESMFoldDB. Feedback welcome :)
bioicons.com/pdb2vector/
Michael Stam retweeted
#333: Ft. @elmtownpodcast & @hayleighdotdev, @ChrisWellsWood & @mjstam & @MartaChronowska & @WoolfsonLab, @nelsonic, and @7h3_360l355_d3v. elmweekly.nl/p/elm-weekly-is… #elmlang
Michael Stam retweeted
Just putting some papers on the timeline:) Looking forward to seeing if we are now in the fourth phase (with deep-learning adding even more designs per year).
Also nice to put some numbers onto things people have long suspected (like overrepresentation of alpha helices).
This Post is from an account that no longer exists. Learn more
Ever wish there was a way to easily browse designed proteins? You're in luck! Introducing the Protein Design Archive - your go-to resource for exploring designed protein structures across the last 40 years!
Website: pragmaticproteindesign.bio.e…
Preprint:
biorxiv.org/content/10.1101/…
This Post is from an account that no longer exists. Learn more
Michael Stam retweeted
A trend we've been seeing in recent de novo protein backbone design is the overrepresentation of alpha helices, compared to very few topologies with beta sheets. Thanks to the resource below, we can see that this has actually been ongoing for decades now
ALT "... in the time periods 2015 – 2020 and 2020 – 2025 the proportion of α helices in designs is 52.5% and 53.5%, respectively, whereas in the PDB it is 32.5%, while in 2015 – 2020 the proportion of β strands is 12.3%, compared to 15.5% in 2020 – 2025 and 21.4% in the PDB."
The Protein Design Archive (PDA): insights from 40 years of protein design
- The Protein Design Archive (PDA) is a comprehensive resource, compiling 40 years of designed proteins from various methodologies. It includes both sequence- and structure-based data.
- One of the key highlights of the PDA is its ability to systematically reveal trends in amino acid usage and secondary structure content in designed proteins compared to natural ones.
- The PDA provides insights into the exponential growth of protein designs, especially in the last decade, driven by deep learning-based approaches in protein engineering.
- The repository features over 1,450 protein designs, showcasing a rich diversity of synthetic proteins. It is freely available for researchers and includes advanced filtering for detailed exploration.
- Analysis within the PDA shows biases in designed proteins, such as overrepresentation of α-helical structures, compared to natural proteins. These insights are vital for refining future design strategies.
- The PDA serves as a valuable tool for the design community, helping users understand the evolution of the field and identifying gaps and opportunities in synthetic biology and bioengineering.
@ChrisWellsWood
💻Website: pragmaticproteindesign.bio.e…
💻Code: github.com/wells-wood-resear…
📜Paper: doi.org/10.1101/2024.09.05.6…
