Sabitlenmiş Tweet
RL finetuning for protein binder design is being hailed as the next obvious step. But I ran the numbers, and something uncomfortable is happening under the hood. 🧵
English
multimodali
200 posts
multimodali
@multimodali
Protein Engineering @harvardmed & @VICatMGH | having a fun time @ucdavis
Davis, CA Katılım Temmuz 2022
807 Takip Edilen223 Takipçiler
English
@owl_posting @zetalyrae As people get inundated with AI content, they'll start to value the effort needed to create something rather than the quality of the creation.
English
I don't want to read any blog post that has an AI-generated illustration.
English
@elyasbuilds wonder if it can detect jailbroken Claude's works? cc'ing @lefthanddraft 👀
English
Just pushed a new blog post: does Pangram work on GPT-5.4?
TLDR, yes!
As with most model releases, Pangram generalizes plenty well to GPT-5.4, without needing to train directly on it's output.
English
multimodali retweetledi
introducing tubestack: bringing substack’s zen to youtube
a quiet, local desktop app to explore youtube without visual pollution. download here for free on mac & windows: anishlk.com/tubestack
(typography taken from the pretty blogs of @AnthropicAI, @OpenAI, @X, @Substack)
English
7/ Takeaway:
Main takeaway: reward gains are useful, but they shouldn’t be read alone.
For RL protein design, composition controls, off-target scores, specificity margins, and independent checks make the biological claim much clearer.
Full writeup:
auralie.substack.com/p/a-friend-to-…
English
6/7 This is where things got interesting.
Stress-granule RL sequences scored highly for stress granule, but also scored highly for P-body.
So the target reward improved, but ProtGPS itself didn’t cleanly certify fine-grained specificity.
English
We set out to audit the RL results for IDiom, a protein language model for designing intrinsically disordered regions.
The question was simple:
Did RL learn specific localization, or did it learn sequences that score broadly well for related compartments?
The results 🧵
Grant Rotskoff@grantrotskoff
Protein design has been dominated by diffusions due to a "structure-first" perspective. What about intrinsically disordered proteins? We scale language-based design using the modern RL stack and our model IDiom. Paper: biorxiv.org/content/10.648… Try it: idiom-designer.vercel.app
English
multimodali retweetledi
"The Bitter Lesson has fully arrived in sequence biology and protein structure. Evo 2, AlphaFold 2 and 3, ProGen3, RFdiffusion".
This sentence has some issues IMO. 1/
Sylvain Gariel@SylvainGariel
English
multimodali retweetledi
Excited to share our new preprint:
“Computational design of membrane fusion proteins”
Huge thanks to all collaborators, co-authors, @KingLabIPD, and everyone at @UWproteindesign who contributed to this work.
Preprint:
biorxiv.org/content/10.648…
English
@aaronmring @xrtcell never in my life did i think i’d see these words 😭😭
English
@xrtcell We are REAPmaxxing my friend! Or REAPmogging, depending on who you ask.
English
If you haven’t REAP’d, have you really lived?
Aaron Ring@aaronmring
Autoantibodies cause autoimmune disease, shape infection outcomes, and alter cancer immunotherapy. But what role might they play in neuropsychiatric disease? In our new preprint, Katlyn Nemani and @JillianRJaycox take on this question in schizophrenia. 🧵 biorxiv.org/content/10.648…
English
multimodali retweetledi
Hiring one intern in the next 48 hours. Requirements:
- Use agents
- Currently in university
- Have coded with GPUs
- Can start now with free time over the next week
Science, computing infra, or cloud is a plus. Will involve biology. Apply by replying with your GitHub username
English
multimodali retweetledi
i think we are all wondering what goes through the minds of the admissions committee period
༗@yetzirot
I know this topic’s been discussed before but I really wonder what goes through the minds of the admissions committee when they see kids like this
English
multimodali retweetledi
Our paper STRIDE was accepted to ICML 2026!
We post-train LLMs to optimize proteins & molecules by emitting a chain-of-thought of atomic edits (INSERT / DELETE / REPLACE). Levenshtein-shortest-path SFT + GRPO-style RL. Boosts protein optimization success from 42% → 89%.
English