[English] 日本語
![](img/lk-miru.gif)
- PDB-8f54: Top-down design of protein architectures with reinforcement learning -
+
Open data
-
Basic information
Entry | Database: PDB / ID: 8f54 | ||||||
---|---|---|---|---|---|---|---|
Title | Top-down design of protein architectures with reinforcement learning | ||||||
![]() | RC_I_1 | ||||||
![]() | DE NOVO PROTEIN / nanoparticle / capsid / oligomer / de novo design / rosetta / reinforcement learning | ||||||
Biological species | synthetic construct (others) | ||||||
Method | ELECTRON MICROSCOPY / single particle reconstruction / cryo EM / Resolution: 2.5 Å | ||||||
![]() | Borst, A.J. / Baker, D. | ||||||
Funding support | ![]()
| ||||||
![]() | ![]() Title: Top-down design of protein architectures with reinforcement learning. Authors: Isaac D Lutz / Shunzhi Wang / Christoffer Norn / Alexis Courbet / Andrew J Borst / Yan Ting Zhao / Annie Dosey / Longxing Cao / Jinwei Xu / Elizabeth M Leaf / Catherine Treichel / Patrisia ...Authors: Isaac D Lutz / Shunzhi Wang / Christoffer Norn / Alexis Courbet / Andrew J Borst / Yan Ting Zhao / Annie Dosey / Longxing Cao / Jinwei Xu / Elizabeth M Leaf / Catherine Treichel / Patrisia Litvicov / Zhe Li / Alexander D Goodson / Paula Rivera-Sánchez / Ana-Maria Bratovianu / Minkyung Baek / Neil P King / Hannele Ruohola-Baker / David Baker / ![]() ![]() ![]() ![]() Abstract: As a result of evolutionary selection, the subunits of naturally occurring protein assemblies often fit together with substantial shape complementarity to generate architectures optimal for function ...As a result of evolutionary selection, the subunits of naturally occurring protein assemblies often fit together with substantial shape complementarity to generate architectures optimal for function in a manner not achievable by current design approaches. We describe a "top-down" reinforcement learning-based design approach that solves this problem using Monte Carlo tree search to sample protein conformers in the context of an overall architecture and specified functional constraints. Cryo-electron microscopy structures of the designed disk-shaped nanopores and ultracompact icosahedra are very close to the computational models. The icosohedra enable very-high-density display of immunogens and signaling molecules, which potentiates vaccine response and angiogenesis induction. Our approach enables the top-down design of complex protein nanomaterials with desired system properties and demonstrates the power of reinforcement learning in protein design. | ||||||
History |
|
-
Structure visualization
Structure viewer | Molecule: ![]() ![]() |
---|
-
Downloads & links
-
Download
PDBx/mmCIF format | ![]() | 1.3 MB | Display | ![]() |
---|---|---|---|---|
PDB format | ![]() | 1.1 MB | Display | ![]() |
PDBx/mmJSON format | ![]() | Tree view | ![]() | |
Others | ![]() |
-Validation report
Arichive directory | ![]() ![]() | HTTPS FTP |
---|
-Related structure data
Related structure data | ![]() 28860MC ![]() 8f4xC ![]() 8f53C C: citing same article ( M: map data used to model this data |
---|
-
Links
-
Assembly
Deposited unit | ![]()
|
---|---|
1 |
|
-
Components
#1: Protein | Mass: 7828.097 Da / Num. of mol.: 60 Source method: isolated from a genetically manipulated source Source: (gene. exp.) synthetic construct (others) / Production host: ![]() ![]() |
---|
-Experimental details
-Experiment
Experiment | Method: ELECTRON MICROSCOPY |
---|---|
EM experiment | Aggregation state: PARTICLE / 3D reconstruction method: single particle reconstruction |
-
Sample preparation
Component | Name: RC_I_1 / Type: COMPLEX / Entity ID: all / Source: RECOMBINANT |
---|---|
Source (natural) | Organism: ![]() ![]() |
Source (recombinant) | Organism: ![]() ![]() |
Buffer solution | pH: 7.5 |
Specimen | Conc.: 1 mg/ml / Embedding applied: NO / Shadowing applied: NO / Staining applied: NO / Vitrification applied: YES |
Specimen support | Grid type: Quantifoil R2/2 |
Vitrification | Cryogen name: ETHANE |
-
Electron microscopy imaging
Experimental equipment | ![]() Model: Titan Krios / Image courtesy: FEI Company |
---|---|
Microscopy | Model: TFS KRIOS |
Electron gun | Electron source: ![]() |
Electron lens | Mode: BRIGHT FIELD / Nominal defocus max: 1700 nm / Nominal defocus min: 800 nm |
Image recording | Electron dose: 61.155 e/Å2 / Detector mode: COUNTING / Film or detector model: GATAN K3 (6k x 4k) |
-
Processing
Software | Name: PHENIX / Version: 1.20.1_4487: / Classification: refinement | ||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
EM software |
| ||||||||||||||||||||||||||||||||||||||||||||||||
CTF correction | Type: NONE | ||||||||||||||||||||||||||||||||||||||||||||||||
Particle selection | Num. of particles selected: 730457 | ||||||||||||||||||||||||||||||||||||||||||||||||
3D reconstruction | Resolution: 2.5 Å / Resolution method: FSC 0.143 CUT-OFF / Num. of particles: 678727 / Symmetry type: POINT | ||||||||||||||||||||||||||||||||||||||||||||||||
Refine LS restraints |
|