Pascal Notin

444 posts

Pascal Notin banner
Pascal Notin

Pascal Notin

@NotinPascal

Research in AI for Protein Design @Harvard | Prev. CS PhD @UniofOxford, Maths & Physics @Polytechnique

Boston Katılım Eylül 2020
406 Takip Edilen1.4K Takipçiler
Sabitlenmiş Tweet
Pascal Notin
Pascal Notin@NotinPascal·
🧬 😴 TIRED: Scaling protein models to billions of parameters hoping they'll memorize all of evolution and generalize beyond 🔥 WIRED: Smart retrieval-augmented models that dynamically access what they need from sequence databases
Ruben Weitzman@ruben_weitzman

🚨ICML Paper Alert🚨 What if finding the right protein homologs wasn't a slow search, but a learned part of the model itself? We introduce 𝐏𝐫𝐨𝐭𝐫𝐢𝐞𝐯𝐞𝐫, an end-to-end framework that learns to retrieve the most useful homologs for self-supervised reconstruction! (1/12)

English
1
5
91
7K
Pascal Notin retweetledi
Jude Wells
Jude Wells@_judewells·
Quite pleased to hear that one of my submitted proteins placed #5 out of 1200 that were tested in this lovely competition. The approach: pure rational design. Nice to see that human + Microsoft Word is still competitive with state-of-the-art AI methods 🙂
Jude Wells tweet media
Adaptyv Bio@adaptyvbio

The results of the Nipah Protein Design Competition are out! 🧬 1200 proteins experimentally validated (3x more than last year) 📈 99 novel binders against the target protein (a challenging tetramer with little prior work) 💪 26 single digit nM or better binders, with the best ones at single-digit picomolar affinity! All data now available open-source on Proteinbase! Let's take a look at the results ⬇️

English
18
76
782
170.8K
Pascal Notin retweetledi
Debora Marks
Debora Marks@deboramarks·
New paper “Proteome-wide model for human disease genetics” is now live at Nature Genetics: rdcu.be/eRu7K popEVE (pop.evemodel.org) finds the needles in the haystacks of human genetic variation:
Debora Marks tweet media
English
1
61
178
30.3K
youngsu ko
youngsu ko@youngsuko9·
@AllThingsApx Without having to align and make MSAs, getting the homologous sequences using vector databases / embedding similarities seem like it will be very convenient and fast.
English
1
0
1
86
Kyle Tretina, Ph.D.
Kyle Tretina, Ph.D.@AllThingsApx·
ESM-2. MSA Transformer. PoET. Now, Profluent-E1. Profluent-E1 is a new SOTA retrieval-augmented model that conditions on unaligned homologs, hits a new SOTA at zero-shot fitness & contact prediction benchmarks. Best part imo? It’s a flexible drop-in for ESM-2 or a retrieval-powered expert.
Kyle Tretina, Ph.D. tweet media
English
3
11
84
9.4K
Pascal Notin retweetledi
Yarin
Yarin@yaringal·
Reminder - PhD applications for OATML are now open The first funding deadline is December 2 - candidates interested in developing Bayesian deep learning methodology, applications of ML, AI security, and understanding ML methodology are encouraged to apply More info: oatml.cs.ox.ac.uk/apply.html
English
3
49
179
27.1K
Pascal Notin retweetledi
Alex Tong
Alex Tong@AlexanderTong7·
Thrilled to announce I'm starting as a Principal Investigator at #Aithyra in Vienna! We'll be developing generative models to understand cell biology and design proteins. I'm hiring PhDs, Postdocs, & Visiting Researchers! PhD applications by Sept 10: apply.cemm.at
English
29
39
330
27.1K
Nathan C. Frey
Nathan C. Frey@nc_frey·
After three exciting years @PrescientDesign @genentech, last Friday was my last day. It has been an amazing journey, with an incredible team, and I know there is much more to come from this stellar group of researchers. I am immensely grateful to my team, our collaborators, and to @stephenrra & @GligorijevicV for their mentorship and support over the years. Onto the next adventure!
Nathan C. Frey tweet media
English
21
5
251
18.9K
Pascal Notin
Pascal Notin@NotinPascal·
@houchao1 Congrats @houchao1 ! Wondering if NLL -- as a measure of how a given protein is constrained -- could be used as a quick diagnostic to assess which protein is "easier to evolve" in protein engineering campaigns
English
1
0
0
72
Hou Chao
Hou Chao@houchao1·
8/n 🔬 Interesting finding: proteins with high likelihoods from MSA-based estimates tend to have more deleterious mutations — a pattern not observed for ESM2.
Hou Chao tweet media
English
2
0
5
486
Hou Chao
Hou Chao@houchao1·
We just updated our manuscript "Understanding Language Model Scaling on Protein Fitness Prediction". Where we explained why larger pLMs don’t always perform better on mutation effect prediction. We extended beyond ESM2 to models like ESMC, ESM3, SaProt, and ESM-IF1. #ProteinLM
Hou Chao tweet media
English
4
23
134
11.8K
Boris Power
Boris Power@BorisMPower·
At @OpenAI, we believe that AI can accelerate science and drug discovery. An exciting example is our work with @RetroBiosciences, where a custom model designed improved variants of the Nobel-prize winning Yamanaka proteins. Today we published a closer look at the breakthrough. ⬇️
Boris Power tweet media
English
160
628
3.6K
2.1M
Pascal Notin retweetledi
Noelia Ferruz
Noelia Ferruz@ferruz_noelia·
I’ve thoroughly enjoyed reading two (VERY!) recent papers that model protein sequences by retrieving evolutionary information (dynamically) at inference time, and there's a lot to unpack! [1] arxiv.org/abs/2506.08954 [2] biorxiv.org/content/10.110… (1/n)
English
2
28
185
17.9K
Pascal Notin retweetledi
nature
nature@Nature·
A generative artificial-intelligence tool has designed a synthetic CRISPR system that successfully edits human DNA go.nature.com/3JadkRy
English
35
136
460
53K
Pascal Notin retweetledi
Aviv Spinner
Aviv Spinner@AvivSpinner·
1/5 Biological data is noisy, redundant, and ever-growing. 🗣️ In our new paper (first paper of my post doc!! ⚡️), we track model performance across 14 years of UniRef100 snapshots to ask: how does pLM performance scale with training data?
Aviv Spinner tweet media
English
1
23
105
11.5K
Pascal Notin
Pascal Notin@NotinPascal·
Congratulations to the entire @ProfluentAI team on this incredible milestone! OpenCRISPR-1 represents a paradigm shift - the first AI-designed CRISPR protein to successfully edit human DNA with fewer off-target effects. We're moving from discovery-based to engineered biology. 🧬
Ali Madani@thisismadani

Excited to have our AI research published in @Nature today. Proud of the @ProfluentBio team and the extensive final version available under open-access. OpenCRISPR is a milestone. It's the first successful demonstration of editing the human genome with a molecule fully designed by AI (bonus: we open-sourced it). It's also a ML case study in engineering functional biological systems that extend beyond nature for real societal needs. The broad adoption of OpenCRISPR has fueled us to build an incredible platform. We are the one-stop shop to enable any type of precise genome engineering. Through partnership, our collaborators leverage frontier AI to cure disease, build personalized medicines, and solve fundamental societal challenges. Some selected highlights on new scientific results below 👇

English
1
4
72
5.3K
Pascal Notin retweetledi
The Align Foundation
The Align Foundation@Align_Bio·
1/4 🚀 Announcing the 2025 Protein Engineering Tournament. This year’s challenge: design PETase enzymes, which degrade the type of plastic in bottles. Can AI-guided protein design help solve the climate crisis? Let’s find out! ⬇️ #AIforBiology #ClimateTech #ProteinEngineering #OpenScience
The Align Foundation tweet media
English
6
72
194
41.5K
Pascal Notin retweetledi
Patrick Schwab
Patrick Schwab@schwabpa·
Save the date! Machine Learning for Drug Discovery (MLDD) is happening soon on Monday 30 June, 2025. MLDD aims to bring together ML for drug discovery experts, innovators, and enthusiasts from the machine learning, biotechnology and drug discovery domains in London, UK to converge, exchange ideas, and forge new paths in revolutionizing therapeutic development. This year, we are very fortunate to feature keynotes by: - Charlotte Brunne (EPFL) - John Chodera (Achira / MSKCC) - Kexin Huang (Stanford University) - Jacob Kimmel (NewLimit) - Marinka Zitnik (Harvard University) To accommodate the worldwide community, the event will be held fully remotely this year - a link to join and further information are provided at the MLDD website (link in thread below!) Looking forward to seeing you there!
Patrick Schwab tweet media
English
3
28
157
18.9K