Alex Chu's Avatar

Alex Chu

@alexechu.bsky.social

Deep learning & protein design. Currently designing stuff at Google DeepMind, previously at Stanford with Possu Huang. alexechu.github.io

1,474 Followers  |  104 Following  |  6 Posts  |  Joined: 04.10.2023  |  1.6293

Latest posts by alexechu.bsky.social on Bluesky

Preview
a cartoon says hey everybody an old man 's talking while bart simpson looks on ALT: a cartoon says hey everybody an old man 's talking while bart simpson looks on

Bluetorial: Adventures with tenure-Part 1

18.12.2024 12:25 β€” πŸ‘ 43    πŸ” 10    πŸ’¬ 2    πŸ“Œ 0
Preview
FACS‐Sortable Triple Emulsion Picoreactors for Screening Reactions in Biphasic Environments Biphasic reactions can enhance reactions where the catalyst and substrate have different solvent preferences. This study optimizes triple emulsion picoreactors that encapsulate a biphasic solvent env...

Many industrially-important reactions (e.g. lipase-catalyzed biofuel production) involve substrates poorly soluble in water. How can we screen enzymes for these reactions? Samuel Thompson invented a new 'triple emulsion' platform for biphasic screening (1/n):
onlinelibrary.wiley.com/doi/full/10....

13.12.2024 01:54 β€” πŸ‘ 33    πŸ” 10    πŸ’¬ 1    πŸ“Œ 0

At neurips this week - hmu if you’re around, want to catch up, try out some food, explore, etc!

09.12.2024 22:32 β€” πŸ‘ 5    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Latent diffusion ftw!

06.12.2024 17:48 β€” πŸ‘ 6    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Video thumbnail

Super excited to preprint our work on developing a Biomolecular Emulator (BioEmu): Scalable emulation of protein equilibrium ensembles with generative deep learning from @msftresearch.bsky.social ch AI for Science.

www.biorxiv.org/content/10.1...

06.12.2024 08:38 β€” πŸ‘ 440    πŸ” 147    πŸ’¬ 21    πŸ“Œ 29

Evolution is non-linear, the next step in the evolutionary lineage of the ESM series of models is a sequence only protein language model that sets a new state of the art in representation learning.

Check it out! The 300M and 600M models are open weights, 300M available for commercial usage.

04.12.2024 17:50 β€” πŸ‘ 13    πŸ” 4    πŸ’¬ 0    πŸ“Œ 0
Preview
AI Proteins Announces Research Collaboration and Option Agreement with Bristol Myers Squibb for Miniprotein-Based Therapeutics Valued up to $400M AI Proteins, Inc., a biotechnology company that utilizes computational de novo protein design to create therapeutic miniproteins, today announced that

AI Proteins is excited to partner with @bms-news.bsky.social to develop novel miniprotein-based therapeutics. This collaboration will leverage our cutting-edge AI-driven platform to design and optimize miniproteins with the potential to transform patient outcomes
www.businesswire.com/news/home/20...

03.12.2024 17:30 β€” πŸ‘ 3    πŸ” 2    πŸ’¬ 1    πŸ“Œ 0
Post image

β€œOn a log-log plot, my grandmother fits on a straight line.”
-Physicist Fritz Houtermans

There's a lot of truth to this. log-log plots are often abused and can be very misleading

1/5

03.12.2024 04:41 β€” πŸ‘ 43    πŸ” 13    πŸ’¬ 1    πŸ“Œ 1
Post image

AF3 BEST METHOD followed by cluspro but also some conversion errors

03.12.2024 17:07 β€” πŸ‘ 22    πŸ” 10    πŸ’¬ 2    πŸ“Œ 2

go.bsky.app/BoXpqQp
Computational protein design starter pack. Let me know if I missed you!

17.11.2024 11:55 β€” πŸ‘ 58    πŸ” 14    πŸ’¬ 18    πŸ“Œ 1

Much needed! can I be added too?

28.11.2024 23:08 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Asking the following earnestly: what is the strongest case for GANs standing the "test of time"? Are they important 10 years later in modern ML research? How have they influenced the way we think about generative models today?

27.11.2024 21:04 β€” πŸ‘ 65    πŸ” 5    πŸ’¬ 18    πŸ“Œ 4
Preview
A new golden age of discovery In this essay, we take a tour of how AI is transforming scientific disciplines from genomics to computer science to weather forecasting. Some scientists are training their own AI models, while...

A great new essay on AI for Science from our colleagues here:

deepmind.google/public-polic...

26.11.2024 13:35 β€” πŸ‘ 22    πŸ” 5    πŸ’¬ 0    πŸ“Œ 1

Amazing blog post on flow matching, stunning visuals! It also makes the connection with normalising flows crystal clear. Incredible effort!

27.11.2024 18:31 β€” πŸ‘ 90    πŸ” 15    πŸ’¬ 2    πŸ“Œ 1
Post image

A weekend project from a while back -- this little package (with no dependencies) allows you to interact with pymol remotely.

I use it a lot for my protein design workflows together with @biotite.bsky.social.

Just `pip install pymol-remote`

25.11.2024 14:50 β€” πŸ‘ 47    πŸ” 13    πŸ’¬ 4    πŸ“Œ 3
Preview
Scaling Down for Efficiency: Medium-Sized Transformer Models for Protein Sequence Transfer Learning Protein language models such as the transformer-based Evolutionary Scale Modeling 2 (ESM2) can offer deep insights into evolutionary and structural properties of proteins. While larger models, such as...

The complete guide for transfer learning with the protein language model ESM-2.

(In brief: Use ESM-2 650M and calculate mean embeddings across sites.)
www.biorxiv.org/content/10.1...

25.11.2024 02:50 β€” πŸ‘ 49    πŸ” 21    πŸ’¬ 2    πŸ“Œ 3

Thanks! excited to be here

16.11.2024 03:48 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image Post image Post image Post image

Experimental kcat and KM measurements for hundreds of naturally variants of adenylate kinase.
- thermophilic enzymes are not slower than mesophilic!
- general kcat/KM predictors are bad and easily beaten by models trained on this specific dataset!

www.biorxiv.org/content/10.1...

13.11.2024 22:30 β€” πŸ‘ 16    πŸ” 5    πŸ’¬ 2    πŸ“Œ 1

πŸ‘‹

16.11.2024 03:45 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Preview
Laboratory Research Scientist, Protein Design, Science London, UK

Our team at deepmind is hiring! boards.greenhouse.io/deepmind/job...

14.11.2024 17:52 β€” πŸ‘ 6    πŸ” 2    πŸ’¬ 0    πŸ“Œ 0

@alexechu is following 20 prominent accounts