Jue Wang (@jueseph) / X

Jue Wang

2,984 posts

Jue Wang

@jueseph

Research scientist @googledeepmind. Formerly @UWproteindesign @ginkgo @HMS_SysBio.

London, UK

juewang.strikingly.com

Joined March 2009

Pinned
Jue Wang
@jueseph
Mar 5, 2024
I wrote a review (with @_JosephWatson @sid_thesci_kid ) on protein design using structure prediction models--i.e. hallucination, inpainting, diffusion. This approach to protein design started as a convenience but has turned out to be surprisingly powerful. cshperspectives.cshlp.org/content/early/…
11K
Jue Wang
@jueseph
Aug 29, 2018
Ok this is definitely a worthy submission to the hall of shame. #sdspagen00b
Jue Wang
@jueseph
Jul 22, 2022
Our work on creating de novo proteins with functional motifs (binding interfaces, active sites, etc) using deep learning is published! Since the preprint, we have improved the methods and added experimental validation! 1/8
science.org
Scaffolding protein functional sites using deep learning
Deep-learning methods enable the scaffolding of desired functional residues within a well-folded designed protein.
Jue Wang
@jueseph
Nov 12, 2021
New preprint! We came up with 2 methods to design de novo scaffold proteins to hold arbitrary functional motifs: hallucination (optimizing a seq against predictions of RoseTTAFold (RF)) and inpainting (recover masked regions of seq+struc.) 1/n
biorxiv.org
Deep learning methods for designing proteins scaffolding functional sites
Current approaches to de novo design of proteins harboring a desired binding or catalytic motif require pre-specification of an overall fold or secondary structure composition, and hence considerable...
Jue Wang
@jueseph
Nov 14, 2024
My team at Deepmind (protein design) is hiring an experimentalist with enzyme expertise. Please RT and/or apply! I'm happy to answer any questions as well.
job-boards.greenhouse.io
DeepMind
36K
Jue Wang
@jueseph
Aug 20, 2020
People have been trying to get E. coli to grow on methanol for more than a decade. Liao group finally succeeded. Mix of rational deletions and experimental evolution. A 70-kb genome region adaptively changes copy number based on if medium is MeOH or LB!
Converting Escherichia coli to a Synthetic Methylotroph Growing Solely on Methanol
From cell.com
Jue Wang
@jueseph
Sep 23, 2023
Some backstory behind AF2 that I hadn't known about:
pnas.org
Starting at Go: Protein structure prediction succumbs to machine learning | PNAS
This year’s Albert Lasker Basic Medical Research Award recognizes the invention of AlphaFold, a revolutionary advance in the history of protein res...
12K
Jue Wang
@jueseph
Aug 12, 2022
Lot of (deserved) excitement over protein language-model based structure prediction recently, but there is conflation of "not requiring (explicit) MSA" and "modeling physics of single sequence", e.g. in this commentary. 1/2 sciencedirect.com/science/articl…
Jue Wang
@jueseph
Oct 10, 2021
Replying to @NickCho
Not sure if this actually gets “shut the f* down” in reality. I’ve noticed that some Chinese Americans are all too happy to support exclusionary zoning and all kinds of racially inequitable policy
Jue Wang
@jueseph
Oct 10, 2023
We made RoseTTAFold that predicts everything in the PDB, not just proteins, the used it to make RFDiffusion that generates ligand-binding proteins, including a 10nM binder of digoxigenin. See @r_krishna3's thread for details.
Rohith Krishna
@r_krishna3
Oct 10, 2023
Very excited to share RoseTTAFold All-Atom and RFdiffusion All-Atom, methods for structure prediction and design of biomolecular assemblies! biorxiv.org/content/10.110… 1/n
10K
Jue Wang
@jueseph
Oct 9, 2024
Congratulations to David, John, and @demishassabis! I was inspired to join David's lab in part due to AF1 (and in part due to just David as a person). This is well deserved by all!
The Nobel Prize
@NobelPrize
Oct 9, 2024
BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Chemistry with one half to David Baker “for computational protein design” and the other half jointly to Demis Hassabis and John M. Jumper “for protein structure prediction.”
3K
Jue Wang
@jueseph
Oct 18, 2021
We finally made this code (gradient-based functional motif scaffolding using trRosetta) public, so please check it out. Stay tuned for some newer work along these lines soon! github.com/dtischer/trdes… (with Doug, @sid_thesci_kid @sokrypton, Ivan A.)
Jue Wang
@jueseph
Nov 30, 2020
My first (contribution to a) paper in Baker lab! We take an Alphafold1-like protein structure prediction model and invert it to design de novo proteins with a part of the structure specified and the rest of it "hallucinated". This can be used to create mimics of natural binders.
Jue Wang
@jueseph
Apr 22, 2022
One of the most ambitious projects I've seen in Baker lab. I'm still a bit shocked that it even works. De novo designed protein rotors that spin around designed axles.
Institute for Protein Design
@UWproteindesign
Apr 22, 2022
Computational design of mechanically coupled axle-rotor protein assemblies science.org/doi/10.1126/sc…
Jue Wang
@jueseph
Oct 5, 2023
I want to emphasize the coolest thing about this (IMO) -- using RoseTTAFold inpainting to "clean up" native backbones, replacing long loops with more "designable" helices. This leads to higher stability, expression, and ultimately, function.
Kiera Sumida
@KieraSumida
Oct 4, 2023
Stability issues bringing you down? Try ProteinMPNN! Our preprint describing sequence redesign of natural proteins with ProteinMPNN for improved properties is up. Check it out: biorxiv.org/content/10.110…
11K