Clay Kosonocky
@kosonocky.bsky.social
ML + Biochemistry PhD Candidate at UT Austin. BioML Society Founder. All problems are solvable, so let's solve some
biomlsociety.org
biomlsociety.org
Pinned
Clay Kosonocky
@kosonocky.bsky.social
· Sep 2
🚨The Bits to Binders Competition has concluded!🧬
One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors
Spoiler: They work!
One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors
Spoiler: They work!
🚨The Bits to Binders Competition has concluded!🧬
One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors
Spoiler: They work!
One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors
Spoiler: They work!
September 2, 2025 at 2:45 PM
🚨The Bits to Binders Competition has concluded!🧬
One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors
Spoiler: They work!
One year ago we gathered scientists from around the world to design and submit protein binders that cause immune cells to target and eliminate CD20+ tumors
Spoiler: They work!
For better or worse, I've subconsciously adopted a policy of "AI/fake/bot unless proven innocent" w/ regards to internet content from anyone I don't know personally
May 17, 2025 at 2:18 PM
For better or worse, I've subconsciously adopted a policy of "AI/fake/bot unless proven innocent" w/ regards to internet content from anyone I don't know personally
Some recent used bookstore acquisitions of mine
Part of a few themes I'm trying to understand:
- cultures & life via classic lit
- basics of philosophy & history
- modern political climate & trajectory
- how can we make cities better?
- what will the future look like?
Part of a few themes I'm trying to understand:
- cultures & life via classic lit
- basics of philosophy & history
- modern political climate & trajectory
- how can we make cities better?
- what will the future look like?
May 17, 2025 at 2:15 PM
Some recent used bookstore acquisitions of mine
Part of a few themes I'm trying to understand:
- cultures & life via classic lit
- basics of philosophy & history
- modern political climate & trajectory
- how can we make cities better?
- what will the future look like?
Part of a few themes I'm trying to understand:
- cultures & life via classic lit
- basics of philosophy & history
- modern political climate & trajectory
- how can we make cities better?
- what will the future look like?
Effect of SMILES representation on structure prediction, now with a large sample size!
And we see it more clearly in these plots.
Ideally if there were no effect, the experimental conditions (plots 2 & 3) would be as tight with the y=x as the 1st plot. But we can see that some ligands just don't behave as well
Ideally if there were no effect, the experimental conditions (plots 2 & 3) would be as tight with the y=x as the 1st plot. But we can see that some ligands just don't behave as well
February 12, 2025 at 5:35 PM
Effect of SMILES representation on structure prediction, now with a large sample size!
PSA: Digging into Chai-1 and found out that it gives different protein-ligand structures & confidences depending on the SMILES canonicalization used
May also be the case with AF3 & Boltz-1, but haven't verified
As to what this means... see the rest of the thread 🧵
May also be the case with AF3 & Boltz-1, but haven't verified
As to what this means... see the rest of the thread 🧵
February 4, 2025 at 8:38 PM
PSA: Digging into Chai-1 and found out that it gives different protein-ligand structures & confidences depending on the SMILES canonicalization used
May also be the case with AF3 & Boltz-1, but haven't verified
As to what this means... see the rest of the thread 🧵
May also be the case with AF3 & Boltz-1, but haven't verified
As to what this means... see the rest of the thread 🧵
The DNA for the Bits to Binders competition has just arrived from @twistbio.bsky.social! It is now being handed off to LEAH Labs where they will begin testing the 12,000 binders in a few days 👀
Stay tuned!
Stay tuned!
January 30, 2025 at 5:36 PM
The DNA for the Bits to Binders competition has just arrived from @twistbio.bsky.social! It is now being handed off to LEAH Labs where they will begin testing the 12,000 binders in a few days 👀
Stay tuned!
Stay tuned!
This year I've read 30 books so far. Here are the ~10 that I found the most valuable 🧵
1) One Hundred Years of Solitude (1967) by Gabriel Garcia Marquez
Fantastic novel that I think everybody should read once. Captures the feeling of human life that feels instinctually correct
1) One Hundred Years of Solitude (1967) by Gabriel Garcia Marquez
Fantastic novel that I think everybody should read once. Captures the feeling of human life that feels instinctually correct
December 22, 2024 at 5:13 PM
This year I've read 30 books so far. Here are the ~10 that I found the most valuable 🧵
1) One Hundred Years of Solitude (1967) by Gabriel Garcia Marquez
Fantastic novel that I think everybody should read once. Captures the feeling of human life that feels instinctually correct
1) One Hundred Years of Solitude (1967) by Gabriel Garcia Marquez
Fantastic novel that I think everybody should read once. Captures the feeling of human life that feels instinctually correct
Unparalleled levels of productivity this week
December 20, 2024 at 7:31 PM
Unparalleled levels of productivity this week
@ddelalamo.bsky.social Just looked at the data. We fine-tuned ProteinMPNN on AlphaFold structures from each class of temperature in the HotProtein dataset. We found no change in performance on FireProtDB ddG or dTM.
Work done w/ Phillip Woolley at UT Austin
Work done w/ Phillip Woolley at UT Austin
December 3, 2024 at 4:23 PM
@ddelalamo.bsky.social Just looked at the data. We fine-tuned ProteinMPNN on AlphaFold structures from each class of temperature in the HotProtein dataset. We found no change in performance on FireProtDB ddG or dTM.
Work done w/ Phillip Woolley at UT Austin
Work done w/ Phillip Woolley at UT Austin
Last year I made an intro Protein + ML project to teach BioML Society members about protein embeddings. Here it is! Hope you all enjoy
Should get you familiar extracting embeddings for proteins, visualizing them, and using them for property prediction
Should get you familiar extracting embeddings for proteins, visualizing them, and using them for property prediction
November 22, 2024 at 6:12 PM
Last year I made an intro Protein + ML project to teach BioML Society members about protein embeddings. Here it is! Hope you all enjoy
Should get you familiar extracting embeddings for proteins, visualizing them, and using them for property prediction
Should get you familiar extracting embeddings for proteins, visualizing them, and using them for property prediction
"open-source and commercially available" 😎
Thrilled to announce Boltz-1, the first open-source and commercially available model to achieve AlphaFold3-level accuracy on biomolecular structure prediction! An exciting collaboration with Jeremy, Saro, and an amazing team at MIT and Genesis Therapeutics. A thread!
November 18, 2024 at 1:19 AM
"open-source and commercially available" 😎
Hi new followers. Y'all might find this protein design resource list useful. Lots of great papers + available tools here!
github.com/kosonocky/bi...
github.com/kosonocky/bi...
GitHub - kosonocky/bits-to-binders-resources: Resources for the BioML Challenge 2024: Bits to Binders
Resources for the BioML Challenge 2024: Bits to Binders - kosonocky/bits-to-binders-resources
github.com
November 13, 2024 at 2:38 PM
Hi new followers. Y'all might find this protein design resource list useful. Lots of great papers + available tools here!
github.com/kosonocky/bi...
github.com/kosonocky/bi...
Has anybody been curating a list of the most seminal papers in the BioML space?
Mainly interested in work that expands our ability to do and understand new things rather than improve upon benchmarks
Mainly interested in work that expands our ability to do and understand new things rather than improve upon benchmarks
November 12, 2024 at 2:04 AM
Has anybody been curating a list of the most seminal papers in the BioML space?
Mainly interested in work that expands our ability to do and understand new things rather than improve upon benchmarks
Mainly interested in work that expands our ability to do and understand new things rather than improve upon benchmarks
Reposted by Clay Kosonocky
Anybody have a bioml starter pack?
November 11, 2024 at 11:45 PM