Models
Structural models for use in virtual screening, free energy calculations, and molecular simulation.
Many models come from the Coronavirus Structural Task Force.
Data classification:- Models: Derived, integrated, or refined structures from multiple data sources prepared for different computational tasks.
- Structures: Data defining structures determined by experimental methods and referenced via a unique identifier such as a PDB ID.
- Simulations: The datasets produced as a result of applying the models to different scientific techniques.
- Unpublished data with no preprints have no indicator
- Data in a preprint or submitted for publication are given this
marker. If the preprint is available, it will always show and should work as a link - Data published in a paper or accepted are given this
marker (i.e. has been approved by formal peer review) and should work as links.
Quick Navigation
3CLpro ACE2 BoAT1 E protein Fc receptor Furin Helicase IL6R M protein Macrodomain N protein NSP1 NSP10 NSP11 NSP14 NSP15 NSP16 NSP2 NSP4 NSP6 NSP7 NSP8 NSP9 ORF10 ORF3a ORF6 ORF7a ORF7b ORF8 PD-1 PLpro RdRP TMPRSS2 fusion core p38 spike virion
Models of Virion Particle
Coarse Grained SARS-CoV-2 Virion
Alvin Yu, Gregory A. Voth
University of Chicago -- Voth Lab
Represented Proteins: virion
Model: Files | Source Structure PDBs: ---
A coarse-grained (CG) model for the SARS-CoV-2 virion (in its entirety) developed under an NSF RAPID award 2029092 to G.A.V. in response to the COVID-19 pandemic.
Simulations:
---
Models of Viral Spike Proteins
Viral Spike Fusion Core
SARS-CoV-2 Spike (S) glycoprotein
Blocking SARS-CoV-2 Spike protein binding to human ACE2 receptor
spike_full-length_closed_amarolab
Amaro Lab
University of California San Diego
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX
PSF/PDB for full-length SPIKE protein in the Closed state, including protein, glycans, membrane, water and ions.
Simulations:
Trajectories of full-length SPIKE protein in the Closed state.
Spike protein in complex with human ACE2
Oostenbrink Lab
University of Natural Resources and Life Sciences, Vienna
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: 6VYB 6M17
Atomistic model of the Spike protein in complex with the human ACE2 receptor, most probale glycosylations are added.
Simulations:
Trajectory of the Spike protein in complex with human ACE2
Improved trimeric SARS-CoV-2 spike glycoprotein (closed state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in closed state. The C- and N-peptide termini are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
spike by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: spike
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2
Negin Forouzesh, Alexey Onufriev
California State University, Los Angeles and Virginia Tech
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Truncated structure of SARS-CoV-2 spike receptor-binding domain bound with the human ACE2 receptor.
Simulations:
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model with N343 glycan
William Glass
Represented Proteins: spike RBD
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Takes the RBD (Chain E) from the 6m0j refined model, builds in the N343 glycan using PyMOL, and refines clashes using ISOLDE. See a complete description of the workflow for building and refining the glycosylated model.
Simulations:
Folding@home simulations of the SARS-CoV-2 spike RBD bound to human ACE2
Coarse Grained SPIKE Protein (PDB 2AJF)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: spike
Model: Files | Source Structure PDBs: 2AJF
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 18 mappings of different resolutions are currently available. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
SWISS-MODEL Spike + ACE2
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model
Tristan Croll
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Refinement of 6m0j that fixes multiple issues: * Adds/extend GlcNAcs in ACE2 * Fixes backwards His374 (zinc binding site) * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects rotamer adjustments and peptide flips See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 spike protein trimer (closed state) model for MD simulations
Vladimir Mironov
Lomonosov Moscow State University -- Chemistry Department -- Laboratory of Chemical Cybernetics
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6ACK | Visualize: 3DMol.js
Model of the spike protein trimer of SARS-CoV-2 virus in closed conformation. That trimer is built from the homology model of monomer made by Thomas Desautels et.al. (https://doi.org/10.1101/2020.04.03.024885) It has been replicated and fitted to the known structure of SARS-CoV-1 by Wenfei Song et.al. PDBID: 6ACK (https://doi.org/10.1371/journal.ppat.1007236). That model has been made to perform MD simulations.
Simulations:
Gromacs 60 ns MD of SARS-CoV-2 spike trimer, All Atom model
Trimeric SARS-CoV-2 spike glycoprotein (open state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VYB | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in open state. The C- and N-peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The total number of atoms in the system was 715439.
Simulations:
SWISS-MODEL Spike
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Trimeric SARS-CoV-2 spike glycoprotein (closed state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in closed state. The C- and N-peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The total number of atoms in the system was 566502.
Simulations:
spike_full-length_open_amarolab
Amaro Lab
University of California San Diego
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VSB
PSF/PDB for full-length SPIKE protein in the Open state, including protein, glycans, membrane, water and ions.
Simulations:
Trajectories of full-length SPIKE protein in the Open state (N165A / N234A mutations).
Trajectories of full-length SPIKE protein in the Open state.
SARS-CoV-2 trimeric spike protein binding to FDA approved or investigational drug molecules
DESRES
Represented Proteins: spike RBD
Model: Files | Source Structure PDBs: 6VXX 6VW1 | Visualize: 3DMol.js
Spike trimer constructed from PDB entries 6VXX and 6VW1, only retaining the RBD and a short region from S1 fusion protein as a minimal system for maintaining a trimer assembly. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns. The simulations were conducted at 310 K in the NPT ensemble.
Simulations:
Inhibiting cleavage of the SARS-CoV-2 spike protein
Trimeric SARS-CoV-2 spike glycoprotein (open state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VYB | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in open state. The C- and N-peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The total number of atoms in the system was 715439.
Simulations:
SWISS-MODEL Spike
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Trimeric SARS-CoV-2 spike glycoprotein (closed state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in closed state. The C- and N-peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The total number of atoms in the system was 566502.
Simulations:
SARS-CoV-2 trimeric spike protein binding to FDA approved or investigational drug molecules
DESRES
Represented Proteins: spike RBD
Model: Files | Source Structure PDBs: 6VXX 6VW1 | Visualize: 3DMol.js
Spike trimer constructed from PDB entries 6VXX and 6VW1, only retaining the RBD and a short region from S1 fusion protein as a minimal system for maintaining a trimer assembly. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns. The simulations were conducted at 310 K in the NPT ensemble.
Simulations:
spike_full-length_open_amarolab
Amaro Lab
University of California San Diego
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VSB
PSF/PDB for full-length SPIKE protein in the Open state, including protein, glycans, membrane, water and ions.
Simulations:
Trajectories of full-length SPIKE protein in the Open state.
Trajectories of full-length SPIKE protein in the Open state (N165A / N234A mutations).
spike_full-length_closed_amarolab
Amaro Lab
University of California San Diego
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX
PSF/PDB for full-length SPIKE protein in the Closed state, including protein, glycans, membrane, water and ions.
Simulations:
Trajectories of full-length SPIKE protein in the Closed state.
Spike protein in complex with human ACE2
Oostenbrink Lab
University of Natural Resources and Life Sciences, Vienna
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: 6VYB 6M17
Atomistic model of the Spike protein in complex with the human ACE2 receptor, most probale glycosylations are added.
Simulations:
Trajectory of the Spike protein in complex with human ACE2
Improved trimeric SARS-CoV-2 spike glycoprotein (closed state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in closed state. The C- and N-peptide termini are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
spike by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: spike
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2
Negin Forouzesh, Alexey Onufriev
California State University, Los Angeles and Virginia Tech
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Truncated structure of SARS-CoV-2 spike receptor-binding domain bound with the human ACE2 receptor.
Simulations:
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model with N343 glycan
William Glass
Represented Proteins: spike RBD
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Takes the RBD (Chain E) from the 6m0j refined model, builds in the N343 glycan using PyMOL, and refines clashes using ISOLDE. See a complete description of the workflow for building and refining the glycosylated model.
Simulations:
Folding@home simulations of the SARS-CoV-2 spike RBD bound to human ACE2
Coarse Grained SPIKE Protein (PDB 2AJF)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: spike
Model: Files | Source Structure PDBs: 2AJF
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 18 mappings of different resolutions are currently available. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
SWISS-MODEL Spike + ACE2
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model
Tristan Croll
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Refinement of 6m0j that fixes multiple issues: * Adds/extend GlcNAcs in ACE2 * Fixes backwards His374 (zinc binding site) * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects rotamer adjustments and peptide flips See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 spike protein trimer (closed state) model for MD simulations
Vladimir Mironov
Lomonosov Moscow State University -- Chemistry Department -- Laboratory of Chemical Cybernetics
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6ACK | Visualize: 3DMol.js
Model of the spike protein trimer of SARS-CoV-2 virus in closed conformation. That trimer is built from the homology model of monomer made by Thomas Desautels et.al. (https://doi.org/10.1101/2020.04.03.024885) It has been replicated and fitted to the known structure of SARS-CoV-1 by Wenfei Song et.al. PDBID: 6ACK (https://doi.org/10.1371/journal.ppat.1007236). That model has been made to perform MD simulations.
Simulations:
Gromacs 60 ns MD of SARS-CoV-2 spike trimer, All Atom model
Inhibition of formation of the viral fusion core
spike by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: spike
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2
Negin Forouzesh, Alexey Onufriev
California State University, Los Angeles and Virginia Tech
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Truncated structure of SARS-CoV-2 spike receptor-binding domain bound with the human ACE2 receptor.
Simulations:
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model with N343 glycan
William Glass
Represented Proteins: spike RBD
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Takes the RBD (Chain E) from the 6m0j refined model, builds in the N343 glycan using PyMOL, and refines clashes using ISOLDE. See a complete description of the workflow for building and refining the glycosylated model.
Simulations:
Folding@home simulations of the SARS-CoV-2 spike RBD bound to human ACE2
Coarse Grained SPIKE Protein (PDB 2AJF)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: spike
Model: Files | Source Structure PDBs: 2AJF
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 18 mappings of different resolutions are currently available. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
SWISS-MODEL Spike + ACE2
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model
Tristan Croll
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Refinement of 6m0j that fixes multiple issues: * Adds/extend GlcNAcs in ACE2 * Fixes backwards His374 (zinc binding site) * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects rotamer adjustments and peptide flips See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 spike protein trimer (closed state) model for MD simulations
Vladimir Mironov
Lomonosov Moscow State University -- Chemistry Department -- Laboratory of Chemical Cybernetics
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6ACK | Visualize: 3DMol.js
Model of the spike protein trimer of SARS-CoV-2 virus in closed conformation. That trimer is built from the homology model of monomer made by Thomas Desautels et.al. (https://doi.org/10.1101/2020.04.03.024885) It has been replicated and fitted to the known structure of SARS-CoV-1 by Wenfei Song et.al. PDBID: 6ACK (https://doi.org/10.1371/journal.ppat.1007236). That model has been made to perform MD simulations.
Simulations:
Gromacs 60 ns MD of SARS-CoV-2 spike trimer, All Atom model
Trimeric SARS-CoV-2 spike glycoprotein (open state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VYB | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in open state. The C- and N-peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The total number of atoms in the system was 715439.
Simulations:
SWISS-MODEL Spike
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Trimeric SARS-CoV-2 spike glycoprotein (closed state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in closed state. The C- and N-peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The total number of atoms in the system was 566502.
Simulations:
spike_full-length_open_amarolab
Amaro Lab
University of California San Diego
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VSB
PSF/PDB for full-length SPIKE protein in the Open state, including protein, glycans, membrane, water and ions.
Simulations:
Trajectories of full-length SPIKE protein in the Open state (N165A / N234A mutations).
Trajectories of full-length SPIKE protein in the Open state.
SARS-CoV-2 trimeric spike protein binding to FDA approved or investigational drug molecules
DESRES
Represented Proteins: spike RBD
Model: Files | Source Structure PDBs: 6VXX 6VW1 | Visualize: 3DMol.js
Spike trimer constructed from PDB entries 6VXX and 6VW1, only retaining the RBD and a short region from S1 fusion protein as a minimal system for maintaining a trimer assembly. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns. The simulations were conducted at 310 K in the NPT ensemble.
Simulations:
spike_full-length_closed_amarolab
Amaro Lab
University of California San Diego
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX
PSF/PDB for full-length SPIKE protein in the Closed state, including protein, glycans, membrane, water and ions.
Simulations:
Trajectories of full-length SPIKE protein in the Closed state.
Spike protein in complex with human ACE2
Oostenbrink Lab
University of Natural Resources and Life Sciences, Vienna
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: 6VYB 6M17
Atomistic model of the Spike protein in complex with the human ACE2 receptor, most probale glycosylations are added.
Simulations:
Trajectory of the Spike protein in complex with human ACE2
Improved trimeric SARS-CoV-2 spike glycoprotein (closed state) in aqueous solution
DESRES
Represented Proteins: spike
Model: Files | Source Structure PDBs: 6VXX | Visualize: 3DMol.js
Solvated structure of the trimeric SARS-CoV-2 spike glycoprotein in closed state. The C- and N-peptide termini are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
Models of Viral Protease, Polymerase, and Nonstructured Proteins
SARS-CoV-2 main protease (3CLpro or NSP5)
3CLpro / Mpro activity
SARS-CoV-2 dimeric main protease without ligand based on PDB 6LU7
Teruhisa S. Komatsu, Yohei M. Koyama, Noriaki Okimoto, Gentaro Morimoto, Yousuke Ohno, Makoto Taiji
Riken Biosystems Dynamics Research
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6LU7
Simulations:
Riken BDR 10 Microsecond Trajectory System Snapshot every 10ns
Riken BDR 10 Microsecond Trajectory Protein Snapshot every 200ps
Riken BDR 10 Microsecond Trajectory Protein Snapshot every 1ns
Coarse Grained 6y2f (PDB 6Y2F)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6Y2F
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 9 mappings of different resolutions are currently available. These are generated from the original all-atom PDB structure.
Simulations:
---
Coarse Grained Main Protease Structures (PDB 6LU7 ligands removed)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li,Xu Group, University of Rochester (contributor)
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6LU7
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 15 mappings of different resolutions are currently available. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
Truncated Mpro based on 6Y2F and shape-compliant 3D conformers
P. I. Koukos, M. Réau, A. M. J. J Bonvin
Computational Structural Biology group, Bijvoet Centre for Biomolecular Research, Utrecht University
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6Y2F
Computational docking of the approved subset of Drugbank + active metabolites + investigational compounds of interest against the Main protease (Mpro). The receptor has been truncated by removing residues distant from the binding site to reduce the computational load. 3D conformers were generated from SMILES using OpenEye OMEGA with default settings.
Simulations:
HADDOCK docking of approved Drugbank set against Mpro with a geometric shape model
Coarse Grained 6y2g (PDB 6Y2G)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6Y2G
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 16 mappings of different resolutions are currently available. These are generated from the original all-atom PDB structure.
Simulations:
---
SARS-CoV-2 main protease model for MD simulations
Matt Hurley
Temple University -- Chemistry Department -- Voelz Lab
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Model of the main protease (Mpro/3cl) of SARS-CoV-2 virus.
Structures used for these simulations were obtained using x-ray crystallography from Diamond Light Source via Tim Dudgeon.
A directory of all structures used can be found on our AWS bucket using the AWS CLI:
aws s3 --no-sign-request sync s3://fah-public-data-covid19-absolute-free-energy/receptor_structures .
These conformations were used for all listed simulations pertaining to absolute free energy of protease inhibitors. A detailed database of which receptor was used for each simulation can be found in the dataframe listed in the url field.
Simulations:
SWISS-MODEL 3CLpro
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Truncated Mpro based on 6Y2F and pharmacophore-compliant 3D conformers
P. I. Koukos, M. Réau, A. M. J. J Bonvin
Computational Structural Biology group, Bijvoet Centre for Biomolecular Research, Utrecht University
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6Y2F
Computational docking of the approved subset of Drugbank + active metabolites + investigational compounds of interest against the Main protease (Mpro). The receptor has been truncated by removing residues distant from the binding site to reduce the computational load. 3D conformers were generated from SMILES using OpenEye OMEGA with default settings.
Simulations:
HADDOCK docking of approved Drugbank set against Mpro with a pharmacophore shape model
SARS-CoV-2 main protease (apo, monomer) for Folding@home simulations
Ivy Zhang, Hannah Bruce Macdonald
Folding@home -- Memorial Sloan Kettering Cancer Center -- Chodera lab
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6LU7 | Visualize: 3DMol.js
Apo, monomeric Mpro/3CLpro from PDB structure 6lu7 used in Folding@home simulations. Chain A (i.e. a monomer of the protein, without the inhibitor or waters) was extracted from 6lu7 using PyMOL, and protonated and capped (ACE, NME) with Schrodinger’s Maestro. The protein was solvated in a cubic box with 1 nm padding and 150 mM NaCl using OpenMM 7.4.1.
Simulations:
Folding@home SARS-CoV-2 main protease (apo, monomer) simulations
Coarse Grained COVID-19 main protease in apo form (PDB 6M03)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6M03
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 17 mappings of different resolutions are currently available. These are generated from the original all-atom PDB structure.
Simulations:
---
3CLpro prepared for simulation in a 120 cubic A box for long continuous trajectory
DESRES
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6Y84
Apo enzyme started from the apo enzyme structure determined by X-ray crystallography (PDB entry 6Y84) The protein was solvated in a 120 x 120 x 120 Å water box containing 0.15 M NaCl.
Simulations:
DESRES 100 µs MD of 3CLpro, no water or ions
DESRES 100 µs MD of 3CLpro, All Atom
3CLpro by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Inhibition of viral polymerases
Coarse Grained COVID-19 main protease in apo form (PDB 6M03)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: 3CLpro
Model: Files | Source Structure PDBs: 6M03
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 17 mappings of different resolutions are currently available. These are generated from the original all-atom PDB structure.
Simulations:
---
No Targets Recorded
SARS-CoV-2 Macrodomain (NSP3)
SARS-CoV-2 Papain-like protease (NSP3)
Inhibition of PLpro protease activity
SARS-CoV-1 ligand-bound (PDB 4OW0)
Chang Group
University of California, Riverside
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 4OW0
MD simulations prepared using Amber18 MD package with GPU acceleration. Used FF14SB for the protein, GAFF2 for the ligands. Systems solvated with TIP3p water. Systems minimized in four stages, using Generalized Born implicit solvent to minimize the hydrogen atoms, then the protein sidechains, then the entire protein, and finally the entire solvated structure.
Simulations:
Refinement of SWISS-MODEL PL-Pro domain 2 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
PLpro by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Refinement of RaptorX PLpro by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
SARS-CoV-2 ligand-free (PDB 6W9C - chain C)
Chang Group
University of California, Riverside
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 6W9C
MD simulations prepared using Amber18 MD package with GPU acceleration. Used FF14SB for the protein. Systems solvated with TIP3p water. Systems minimized in four stages, using Generalized Born implicit solvent to minimize the hydrogen atoms, then the protein sidechains, then the entire protein, and finally the entire solvated structure.
Simulations:
Apo SARS-CoV-2 PLPro (from PDB 6W9C C-chain)
Refinement of PLpro by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
SARS-CoV-1 ligand-free (PDB 4OW0 ligand removed)
Chang Group
University of California, Riverside
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 4OW0
MD simulations prepared using Amber18 MD package with GPU acceleration. Used FF14SB for the protein. Systems solvated with TIP3p water. Systems minimized in four stages, using Generalized Born implicit solvent to minimize the hydrogen atoms, then the protein sidechains, then the entire protein, and finally the entire solvated structure.
Simulations:
Coarse Grained Papain-like Protease (PDB 2FE8)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 2FE8
Coarse graied (CG) mappings are generated from the DSGPM deep learning model.. Chains A,B and C are coarse grained separately. There are 24,23 and 24 mappings of different resolutions respectively. These are generated from the original all-atom PDB structure.
Simulations:
---
SARS-CoV-2 PLpro: ISOLDE refined model
Tristan Croll
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 6W9C | Visualize: 3DMol.js
Refinement of 6w9c (a 2.7A X-ray structure) of papain-like protease of SARS-CoV-2 that contains three copies of the same chain in the asymmetric unit. Corrects several issues: * Corrects zinc site in chain A misidentified as a disulfide, and incompletely modeled site in chain B; the same zinc site is correctly modeled in chain C. * Cleans up lots of issues due to missing data (overall completeness 57.1%) See a complete description of the issues remedied by this model.
Simulations:
---
Refinement of SWISS-MODEL PL-Pro domain 7 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coarse Grained Inhibitors of Human Coronavirus Papain-Like Proteases (PDB 4OW0 ligands removed)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 4OW0
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 12 mappings of different resolutions are available currently. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
Refinement of SWISS-MODEL PL-Pro domain 4 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of SWISS-MODEL PL-Pro domain 5 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
SARS-CoV-2 ligand-free (PDB 6WRH)
Chang Group
University of California, Riverside
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 6WRH
MD simulations prepared using Amber18 MD package with GPU acceleration. Used FF14SB for the protein. Systems solvated with TIP3p water. Systems minimized in four stages, using Generalized Born implicit solvent to minimize the hydrogen atoms, then the protein sidechains, then the entire protein, and finally the entire solvated structure.
Simulations:
Apo SARS-CoV-2 PLPro (from PDB 6WRH C-chain)
SARS-CoV-2 ligand-bound (3k ligand was docked to protein conformation from 6W9C ligand-free MD)
Chang Group
University of California, Riverside
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 6W9C
MD simulations prepared using Amber18 MD package with GPU acceleration. Used FF14SB for the protein, GAFF2 for the ligands. Systems solvated with TIP3p water. Systems minimized in four stages, using Generalized Born implicit solvent to minimize the hydrogen atoms, then the protein sidechains, then the entire protein, and finally the entire solvated structure.
Simulations:
3k bound SARS-CoV-2 PLPro (3k docked to frame from trajectory of PDB 6W9C C-chain)
SWISS-MODEL PL-PRO
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Refinement of AplhaFold PLpro C terminal by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Physics-based refinement of Google DeepMind AlphaFold models from early March 2020. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Inhibition of viral polymerases
Coarse Grained Papain-like Protease (PDB 2FE8)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: PLpro
Model: Files | Source Structure PDBs: 2FE8
Coarse graied (CG) mappings are generated from the DSGPM deep learning model.. Chains A,B and C are coarse grained separately. There are 24,23 and 24 mappings of different resolutions respectively. These are generated from the original all-atom PDB structure.
Simulations:
---
No Targets Recorded
SARS-CoV-2 RNA Polymerase (NSP12)
Inhibition of viral polymerases
SWISS-MODEL NSP7 + NSP8 + RdRP
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP7 NSP8 RdRP
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 RdRP (NSP12) in complex with cofactors NSP7 and NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 7BTF | Visualize: 3DMol.js
Refinement of 7btf that fixes misthreading in the vininity of residues 910-920 where residues were shifted within density by 9 residues. See a complete description of the issues remedied by this model.
Simulations:
---
Docking-based repurposing study of approved drugs against truncated RdRp
P. I. Koukos, M. Réau, A. M. J. J Bonvin
Computational Structural Biology group, Bijvoet Centre for Biomolecular Research, Utrecht University
Represented Proteins: RdRP
Model: Files | Source Structure PDBs: 7BV2
Computational docking of the approved subset of Drugbank + active metabolites + investigational compounds of interest against the inhibitor-bound (remdesivir) RdRp. The receptor has been truncated by removing residues distant from the binding site to reduce the computational load. 3D conformers were generated from SMILES using OpenEye OMEGA with default settings.
Simulations:
HADDOCK docking of approved Drugbank set against RdRp
SARS-CoV RdRP (NSP12) in complex with cofactors NSP7 and NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR | Visualize: 3DMol.js
Refinement of 6nur that fixes misthreading in the vininity of residues 910-920 where residues were shifted within density by 9 residues. See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 apo-RdRp complex (nsp12+2*nsp8+nsp7) model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 6M71 7BTF 7BV1 | Visualize: 3DMol.js
Model of the apo-protein form of RdRp complex of the SARS-CoV-2 is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71,7BV1) kindly shared with us by Gao et.al. The fitted models have been equilibriated to perform long MD simulations.
Simulations:
Gromacs 300 ns MD of SARS-CoV-2 apo-RdRp model, All Atom model
RdRP by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: RdRP
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SARS-CoV-2 RdRp complex (nsp12+2*nsp8+nsp7) + RNA template-primer + ATP model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 7BTF 7BV2 6YYT | Visualize: 3DMol.js
Model of the RdRp + RNA + ATP complex of the SARS-CoV-2 with non-covalently bound ATP molecule is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71) kindly shared with us by Gao et.al. Further, the model of RdRp complex with RNA and ATP molecule in Tri-phosphate form is modelled based on comparative fitting with previously known poliovirus and norovirus structres (with NTP molecule in hydrophobic cleft). The protein-RNA complex with Remdesivir shows excellent fit (< 0.7 Å) with the recently published RdRp complex with RNA template-primer (PDB:7BV2, 6YYT). The fitted models have been equilibriated to perform long MD simulations.
Simulations:
Gromacs 100 ns MD of SARS-CoV-2 RdRp + RNA template-primer + ATP model, All Atom model
SARS-CoV-2 RdRp complex (nsp12+2*nsp8+nsp7) + RNA template-primer + RTP (Remdesivir Tri-phosphate) model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 6M71 7BV2 | Visualize: 3DMol.js
Model of the RdRp + RNA + NTP complex of the SARS-CoV-2 with non-covalently bound NTP molecule is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71) kindly shared with us by Gao et.al. Further, the model of RdRp complex with RNA and Remdesivir molecule in Tri-phosphate form is modelled based on comparative fitting with previously known poliovirus and norovirus structres (with NTP molecule in hydrophobic cleft). The protein-RNA complex with Remdesivir shows excellent fit (< 0.5 Å) with the recently published RdRp complex with RNA template-primer and covalently bound Remdesivir in mono-phosphate form (PDB:7BV2, 6YYT) The fitted models have been equilibriated to perform long MD simulations.
Simulations:
SARS-CoV-2 RdRP (NSP12) in complex with NSP7 and two copies of NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71 | Visualize: 3DMol.js
Refinement of 6m71 that fixes multiple issues: * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects 1-2 dozen rotamer adjustments and peptide flips * Models C-terminal domain of chain D (one of the NSP8s) using well-resolved chain B See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 nsp7-nsp8-nsp12 RNA polymerase complex in aqueous solution
DESRES
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71 | Visualize: 3DMol.js
The C- and N-peptide termini capped with amide and acetyl groups respectively. The missing loops in the published structural models were manually built as extended peptide conformation. The missing part of Chain D was built through homology modeling using the structure of SARS-CoV-1 polymerase complex (PDB entry 6NUR). The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
No Targets Recorded
Helicase coronavirus nonstructural protein 13 (NSP13)
Inhibition of nsp13 helicase activity
SWISS-MODEL Helicase
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: Helicase
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Helicase (NSP13) by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: Helicase
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus nonstructural protein 1
No Targets Recorded
Refinement of SWISS-MODEL nsp1 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP1
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of nsp1 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP1
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
NSP1 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP1
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus nonstructural protein 10
No Targets Recorded
SWISS-MODEL NSP10 + NSP14
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP10 NSP14
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SWISS-MODEL NSP10
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP10
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SWISS-MODEL NSP10 + NSP16
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP10 NSP16
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
NSP10 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP10
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus nonstructural protein 11
Coronavirus nonstructural protein 14
No Targets Recorded
SWISS-MODEL NSP10 + NSP14
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP10 NSP14
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
NSP14 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP14
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus nonstructural protein 15
No Targets Recorded
NSP15 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP15
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SWISS-MODEL NSP15
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP15
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Coronavirus nonstructural protein 16
No Targets Recorded
SWISS-MODEL NSP10 + NSP16
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP10 NSP16
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
NSP16 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP16
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus nonstructural protein 2
No Targets Recorded
Refinement of AplhaFold nsp2 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Physics-based refinement of Google DeepMind AlphaFold models from early March 2020. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of nsp2 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
NSP2 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Refinement of Baker nsp2 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of Baker lab’s models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of RaptorX nsp2 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coronavirus nonstructural protein 4
No Targets Recorded
Refinement of NSP4 in membrane environment by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Model re-built with additional information. Refined with consideration of membrane environment. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of nsp4 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of AplhaFold nsp4 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Physics-based refinement of Google DeepMind AlphaFold models from early March 2020. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
NSP4 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SWISS-MODEL NSP4
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Refinement of RaptorX nsp4 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of SWISS-MODEL NSP4 domain 3 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP4
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coronavirus nonstructural protein 6
No Targets Recorded
Refinement of RaptorX nsp6 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of NSP6 in membrane environment by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Used the AlphaFold model, refined with consideration of membrane environment. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of AplhaFold nsp6 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Physics-based refinement of Google DeepMind AlphaFold models from early March 2020. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of nsp6 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: NSP6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
NSP6 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus nonstructural protein 7
Inhibition of viral polymerases
NSP7 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP7
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SARS-CoV-2 RdRP (NSP12) in complex with cofactors NSP7 and NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 7BTF | Visualize: 3DMol.js
Refinement of 7btf that fixes misthreading in the vininity of residues 910-920 where residues were shifted within density by 9 residues. See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV RdRP (NSP12) in complex with cofactors NSP7 and NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR | Visualize: 3DMol.js
Refinement of 6nur that fixes misthreading in the vininity of residues 910-920 where residues were shifted within density by 9 residues. See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 apo-RdRp complex (nsp12+2*nsp8+nsp7) model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 6M71 7BTF 7BV1 | Visualize: 3DMol.js
Model of the apo-protein form of RdRp complex of the SARS-CoV-2 is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71,7BV1) kindly shared with us by Gao et.al. The fitted models have been equilibriated to perform long MD simulations.
Simulations:
Gromacs 300 ns MD of SARS-CoV-2 apo-RdRp model, All Atom model
Coarse Grained RNA-dependent RNA polymerase (PDB 6M71)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71
Coarse graied (CG) mappings are generated from the DSGPM deep learning model.. Molecules NSP 7 (chain C) and NSP 8 (chains A and B) of the RNA-dependent RNA polymerase are coarse grained separately. There are 10 and 15 mappings of different resolutions respectively. These are generated from the original all-atom PDB structure.
Simulations:
---
Coarse Grained Apo NSP7, NSP8 Molecules (PDB 7BV1)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: NSP7 NSP8
Model: Files | Source Structure PDBs: 7BV1
Coarse graied (CG) mappings are generated from DSGPM deep learning model.. Molecules NSP 7 (chain C) and NSP 8 (chains B and D) of the apo nsp12-nsp7-nsp8 complex are coarse grained separately. There are 10 and 19 mappings of different resolutions respectively. These are generated from the original all-atom PDB structure.
Simulations:
---
SWISS-MODEL NSP7 + NSP8
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP7 NSP8
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 RdRP (NSP12) in complex with NSP7 and two copies of NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71 | Visualize: 3DMol.js
Refinement of 6m71 that fixes multiple issues: * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects 1-2 dozen rotamer adjustments and peptide flips * Models C-terminal domain of chain D (one of the NSP8s) using well-resolved chain B See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 RdRp complex (nsp12+2*nsp8+nsp7) + RNA template-primer + ATP model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 7BTF 7BV2 6YYT | Visualize: 3DMol.js
Model of the RdRp + RNA + ATP complex of the SARS-CoV-2 with non-covalently bound ATP molecule is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71) kindly shared with us by Gao et.al. Further, the model of RdRp complex with RNA and ATP molecule in Tri-phosphate form is modelled based on comparative fitting with previously known poliovirus and norovirus structres (with NTP molecule in hydrophobic cleft). The protein-RNA complex with Remdesivir shows excellent fit (< 0.7 Å) with the recently published RdRp complex with RNA template-primer (PDB:7BV2, 6YYT). The fitted models have been equilibriated to perform long MD simulations.
Simulations:
Gromacs 100 ns MD of SARS-CoV-2 RdRp + RNA template-primer + ATP model, All Atom model
SARS-CoV-2 RdRp complex (nsp12+2*nsp8+nsp7) + RNA template-primer + RTP (Remdesivir Tri-phosphate) model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 6M71 7BV2 | Visualize: 3DMol.js
Model of the RdRp + RNA + NTP complex of the SARS-CoV-2 with non-covalently bound NTP molecule is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71) kindly shared with us by Gao et.al. Further, the model of RdRp complex with RNA and Remdesivir molecule in Tri-phosphate form is modelled based on comparative fitting with previously known poliovirus and norovirus structres (with NTP molecule in hydrophobic cleft). The protein-RNA complex with Remdesivir shows excellent fit (< 0.5 Å) with the recently published RdRp complex with RNA template-primer and covalently bound Remdesivir in mono-phosphate form (PDB:7BV2, 6YYT) The fitted models have been equilibriated to perform long MD simulations.
Simulations:
SARS-CoV-2 nsp7-nsp8-nsp12 RNA polymerase complex in aqueous solution
DESRES
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71 | Visualize: 3DMol.js
The C- and N-peptide termini capped with amide and acetyl groups respectively. The missing loops in the published structural models were manually built as extended peptide conformation. The missing part of Chain D was built through homology modeling using the structure of SARS-CoV-1 polymerase complex (PDB entry 6NUR). The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
SWISS-MODEL NSP7 + NSP8 + RdRP
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP7 NSP8 RdRP
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
No Targets Recorded
Coronavirus nonstructural protein 8
Inhibition of viral polymerases
SWISS-MODEL NSP7 + NSP8
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP7 NSP8
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 RdRp complex (nsp12+2*nsp8+nsp7) + RNA template-primer + ATP model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 7BTF 7BV2 6YYT | Visualize: 3DMol.js
Model of the RdRp + RNA + ATP complex of the SARS-CoV-2 with non-covalently bound ATP molecule is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71) kindly shared with us by Gao et.al. Further, the model of RdRp complex with RNA and ATP molecule in Tri-phosphate form is modelled based on comparative fitting with previously known poliovirus and norovirus structres (with NTP molecule in hydrophobic cleft). The protein-RNA complex with Remdesivir shows excellent fit (< 0.7 Å) with the recently published RdRp complex with RNA template-primer (PDB:7BV2, 6YYT). The fitted models have been equilibriated to perform long MD simulations.
Simulations:
Gromacs 100 ns MD of SARS-CoV-2 RdRp + RNA template-primer + ATP model, All Atom model
SARS-CoV-2 RdRp complex (nsp12+2*nsp8+nsp7) + RNA template-primer + RTP (Remdesivir Tri-phosphate) model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 6M71 7BV2 | Visualize: 3DMol.js
Model of the RdRp + RNA + NTP complex of the SARS-CoV-2 with non-covalently bound NTP molecule is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71) kindly shared with us by Gao et.al. Further, the model of RdRp complex with RNA and Remdesivir molecule in Tri-phosphate form is modelled based on comparative fitting with previously known poliovirus and norovirus structres (with NTP molecule in hydrophobic cleft). The protein-RNA complex with Remdesivir shows excellent fit (< 0.5 Å) with the recently published RdRp complex with RNA template-primer and covalently bound Remdesivir in mono-phosphate form (PDB:7BV2, 6YYT) The fitted models have been equilibriated to perform long MD simulations.
Simulations:
SARS-CoV-2 RdRP (NSP12) in complex with NSP7 and two copies of NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71 | Visualize: 3DMol.js
Refinement of 6m71 that fixes multiple issues: * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects 1-2 dozen rotamer adjustments and peptide flips * Models C-terminal domain of chain D (one of the NSP8s) using well-resolved chain B See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 nsp7-nsp8-nsp12 RNA polymerase complex in aqueous solution
DESRES
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71 | Visualize: 3DMol.js
The C- and N-peptide termini capped with amide and acetyl groups respectively. The missing loops in the published structural models were manually built as extended peptide conformation. The missing part of Chain D was built through homology modeling using the structure of SARS-CoV-1 polymerase complex (PDB entry 6NUR). The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
SWISS-MODEL NSP7 + NSP8 + RdRP
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP7 NSP8 RdRP
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 RdRP (NSP12) in complex with cofactors NSP7 and NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 7BTF | Visualize: 3DMol.js
Refinement of 7btf that fixes misthreading in the vininity of residues 910-920 where residues were shifted within density by 9 residues. See a complete description of the issues remedied by this model.
Simulations:
---
NSP8 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP8
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
SARS-CoV RdRP (NSP12) in complex with cofactors NSP7 and NSP8: ISOLDE refined model
Tristan Croll
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR | Visualize: 3DMol.js
Refinement of 6nur that fixes misthreading in the vininity of residues 910-920 where residues were shifted within density by 9 residues. See a complete description of the issues remedied by this model.
Simulations:
---
SARS-CoV-2 apo-RdRp complex (nsp12+2*nsp8+nsp7) model for MD simulations
Vaibhav Modi
University of Jyväskylä -- Department of Chemistry and Nanoscience Center -- Computational Biomolecular Chemistry Group
Represented Proteins: RdRP NSP7 NSP8
Model: Files | Source Structure PDBs: 6NUR 6M71 7BTF 7BV1 | Visualize: 3DMol.js
Model of the apo-protein form of RdRp complex of the SARS-CoV-2 is built using homology modelling with the SARS-CoV-1 RdRp complex (PDB:6NUR) as template structure (https://doi.org/10.1038/s41467-019-10280-3). The modelled structure shows excellent fit (< 0.6 Å) to the SARS-CoV-2 RdRp complex (PDB:6M71,7BV1) kindly shared with us by Gao et.al. The fitted models have been equilibriated to perform long MD simulations.
Simulations:
Gromacs 300 ns MD of SARS-CoV-2 apo-RdRp model, All Atom model
Coarse Grained RNA-dependent RNA polymerase (PDB 6M71)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: NSP7 NSP8
Model: Files | Source Structure PDBs: 6M71
Coarse graied (CG) mappings are generated from the DSGPM deep learning model.. Molecules NSP 7 (chain C) and NSP 8 (chains A and B) of the RNA-dependent RNA polymerase are coarse grained separately. There are 10 and 15 mappings of different resolutions respectively. These are generated from the original all-atom PDB structure.
Simulations:
---
Coarse Grained Apo NSP7, NSP8 Molecules (PDB 7BV1)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: NSP7 NSP8
Model: Files | Source Structure PDBs: 7BV1
Coarse graied (CG) mappings are generated from DSGPM deep learning model.. Molecules NSP 7 (chain C) and NSP 8 (chains B and D) of the apo nsp12-nsp7-nsp8 complex are coarse grained separately. There are 10 and 19 mappings of different resolutions respectively. These are generated from the original all-atom PDB structure.
Simulations:
---
No Targets Recorded
Coronavirus nonstructural protein 9
No Targets Recorded
SWISS-MODEL NSP9
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: NSP9
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
NSP9 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: NSP9
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Models of Viral Open Reading Frame Proteins
Coronavirus Open Reading Frame 10
No Targets Recorded
ORF10 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: ORF10
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Refinement of ORF10 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF10
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of RaptorX ORF10 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF10
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coronavirus Open Reading Frame 3a
No Targets Recorded
Refinement of AplhaFold ORF3a by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF3a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Physics-based refinement of Google DeepMind AlphaFold models from early March 2020. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of RaptorX ORF3a by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF3a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of ORF3a by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF3a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
ORF3a by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: ORF3a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Coronavirus Open Reading Frame 6
No Targets Recorded
Refinement of RaptorX ORF6 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
ORF6 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: ORF6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Refinement of ORF6 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF6
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coronavirus Open Reading Frame 7a
No Targets Recorded
SWISS-MODEL Protein 7a
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: ORF7a
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
Refinement of SWISS-MODEL ORF7a by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF7a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
ORF7a by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: ORF7a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Refinement of ORF7a by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF7a
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coronavirus Open Reading Frame 7b
No Targets Recorded
Refinement of RaptorX ORF7b by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF7b
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of ORF7b by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF7b
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Coronavirus Open Reading Frame 8
No Targets Recorded
Refinement of RaptorX ORF8 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF8
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of ORF8 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF8
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
ORF8 by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: ORF8
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Refinement of SWISS-MODEL ORF8 by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: ORF8
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of SWISS-MODEL models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Models of Viral Membrane Proteins
Membrane Glycoprotein
No Targets Recorded
Refinement of M_protein by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: M protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Contact-based structure prediction using trRosetta. 10 models generated and the best scoring was refined through molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of AplhaFold M_protein by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: M protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Physics-based refinement of Google DeepMind AlphaFold models from early March 2020. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of RaptorX M_protein by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: M protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
FeigLab refinement of RaptorX-Contact models. Refinement performed using a protocol based on molecular dynamics (MD) simulations. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
Refinement of M Protein in membrane environment by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: M protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Monomer structure revised using FeigLab and AlphaFold models. Refined with consideration of membrane environment. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
M protein by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: M protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Models of Viral Envelope Proteins
Envelope small membrane protein
No Targets Recorded
Refinement of E protein in membrane environment by FeigLab
Feig Computational Biophysics Lab
Michigan State University
Represented Proteins: E protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
Monomer structure produced using FeigLab pipeline. Refined with consideration of membrane environment. See https://github.com/feiglab/sars-cov-2-proteins for a complete description of the process.
Simulations:
---
E protein by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: E protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Models of Viral Nucleocapsid Proteins
Nucleoprotein
No Targets Recorded
SWISS-MODEL N Protein
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: N protein
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
N protein by Yang Zhang lab
Chengxin Zhang, Wei Zheng, Xiaoqiang Huang, Eric W. Bell, Xiaogen Zhou, Yang Zhang
Depeartment of Computational Medicine and Bioinformatics -- University of Michigan -- Yang Zhang
Represented Proteins: N protein
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
model from deep learning contact-map assisted C-I-TASSER structure simulation
Simulations:
---
Models of Host Proteins
Angiotensin-converting enzyme 2 (ACE2)
Blocking SARS-CoV-2 Spike protein binding to human ACE2 receptor
Native Human Angiotensin Converting Enzyme-Related Carboxypeptidase (ACE2): ISOLDE refined model
Tristan Croll
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42 | Visualize: 3DMol.js
Refinement of 1r42 that fixes multiple issues: * Merges in missing C-terminal domain from 6m17 See a complete description of the issues remedied by this model.
Simulations:
---
Chimeric RBD in complex with human ACE2
DESRES
Represented Proteins: ACE2 RBD
Model: Files | Source Structure PDBs: 6VW1 | Visualize: 3DMol.js
The complex model was solvated in a ~140 Å box of 200 mM NaCl and water, and parameterized with the DES-Amber protein and ion force field, the TIP4P-D water model, and an in-house force field derived from GAFF.
Simulations:
---
Coarse Grained RBD/ACE2-B0AT1 complex (PDB 6M17)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: RBD ACE2
Model: Files | Source Structure PDBs: 6M17
Coarse graied (CG) mappings are generated from the DSGPM deep learning model.. Chains A,C,E and F of the RBD/ACE2-B0AT1 complex are coarse grained separately. There are 16,16,12 and 12 mappings of different resolutions respectively. These were generated from the original all-atom PDB structure.
Simulations:
---
Human ACE2 ectodomain in aqueous solution (apo open state)
DESRES
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42 | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
DESRES-ANTON-10875753 10 µs simulation trajectory of the human ACE2 ectodomain, no water or ions
DESRES-ANTON-10875753 10 µs simulation trajectory of the human ACE2 ectodomain in aqueous solution
Spike protein in complex with human ACE2
Oostenbrink Lab
University of Natural Resources and Life Sciences, Vienna
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: 6VYB 6M17
Atomistic model of the Spike protein in complex with the human ACE2 receptor, most probale glycosylations are added.
Simulations:
Trajectory of the Spike protein in complex with human ACE2
Structure of SARS coronavirus spike receptor-binding domain complexed with its receptor in aqueous solution
DESRES
Represented Proteins: ACE2 RBD
Model: Files | Source Structure PDBs: 2AJF | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns.
Simulations:
A 10 µs simulation of a SARS-CoV-1 and SARS-CoV-2 chimera-ACE2 complex, no water or ions
Coarse Grained 1r4l Structures (PDB 1R4L ligand removed)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R4L
Coarse graied (CG) mappings prepared with DSGPM deep learning model. There are 20 mappings of different resolutions generated from the original all-atom PDB structure.
Simulations:
---
Native Human Angiotensin Converting Enzyme-Related Carboxypeptidase (ACE2): ISOLDE refined model with glycans
William Glass
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42 | Visualize: 3DMol.js
Takes the ACE2 (Chain A) from the 1r42 refined model, builds all glycans using PyMOL, and refines clashes using ISOLDE. See a complete description of the workflow for building and refining the glycosylated model.
Simulations:
Folding@home simulations of the SARS-CoV-2 spike RBD bound to human ACE2
Human ACE2 ectodomain in aqueous solution (inhibitor-bound closed state)
DESRES
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R4L | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
DESRES-ANTON-10875754 10 µs simulation trajectory of the human ACE2 ectodomain in aqueous solution
DESRES-ANTON-10875754 10 µs simulation trajectory of the human ACE2 ectodomain, no water or ions
Coarse Grained ACE2 (PDB 1R42 ligands removed)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 20 mappings of different resolutions are currently available. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
Chimeric RBD in complex with human ACE2
DESRES
Represented Proteins: ACE2 RBD
Model: Files | Source Structure PDBs: 6VW1 | Visualize: 3DMol.js
The complex model was solvated in a ~140 Å box of 200 mM NaCl and water, and parameterized with the DES-Amber protein and ion force field, the TIP4P-D water model, and an in-house force field derived from GAFF.
Simulations:
---
Docking-based repurposing study of approved drugs against truncated human ACE2 ectodomain (inhibitor-bound closed state)
P. I. Koukos, M. Réau, A. M. J. J Bonvin
Computational Structural Biology group, Bijvoet Centre for Biomolecular Research, Utrecht University
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R4L
Computational docking of the approved subset of Drugbank + active metabolites + investigational compounds of interest against the inhibitor-bound (closed-state) ectodomain of hACE2. The receptor has been truncated by removing residues distant from the binding site to reduce the computational load. 3D conformers were generated from SMILES using OpenEye OMEGA with default settings.
Simulations:
HADDOCK docking of approved Drugbank set against human ACE2 ectodomain
SARS-CoV-2 spike receptor-binding domain bound with ACE2
Negin Forouzesh, Alexey Onufriev
California State University, Los Angeles and Virginia Tech
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Truncated structure of SARS-CoV-2 spike receptor-binding domain bound with the human ACE2 receptor.
Simulations:
SARS-CoV-2 RBD/ACE2-B0AT1 complex in aqueous solution
DESRES
Represented Proteins: ACE2 RBD BoAT1
Model: Files | Source Structure PDBs: 6M17 | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns.
Simulations:
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex in aqueous solution
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex, no water or ions
SWISS-MODEL Spike + ACE2
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model
Tristan Croll
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Refinement of 6m0j that fixes multiple issues: * Adds/extend GlcNAcs in ACE2 * Fixes backwards His374 (zinc binding site) * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects rotamer adjustments and peptide flips See a complete description of the issues remedied by this model.
Simulations:
---
Inhibiting cleavage of the SARS-CoV-2 spike protein
SARS-CoV-2 spike receptor-binding domain bound with ACE2
Negin Forouzesh, Alexey Onufriev
California State University, Los Angeles and Virginia Tech
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Truncated structure of SARS-CoV-2 spike receptor-binding domain bound with the human ACE2 receptor.
Simulations:
SWISS-MODEL Spike + ACE2
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model
Tristan Croll
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Refinement of 6m0j that fixes multiple issues: * Adds/extend GlcNAcs in ACE2 * Fixes backwards His374 (zinc binding site) * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects rotamer adjustments and peptide flips See a complete description of the issues remedied by this model.
Simulations:
---
Spike protein in complex with human ACE2
Oostenbrink Lab
University of Natural Resources and Life Sciences, Vienna
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: 6VYB 6M17
Atomistic model of the Spike protein in complex with the human ACE2 receptor, most probale glycosylations are added.
Simulations:
Trajectory of the Spike protein in complex with human ACE2
Inhibition of formation of the viral fusion core
Native Human Angiotensin Converting Enzyme-Related Carboxypeptidase (ACE2): ISOLDE refined model
Tristan Croll
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42 | Visualize: 3DMol.js
Refinement of 1r42 that fixes multiple issues: * Merges in missing C-terminal domain from 6m17 See a complete description of the issues remedied by this model.
Simulations:
---
Chimeric RBD in complex with human ACE2
DESRES
Represented Proteins: ACE2 RBD
Model: Files | Source Structure PDBs: 6VW1 | Visualize: 3DMol.js
The complex model was solvated in a ~140 Å box of 200 mM NaCl and water, and parameterized with the DES-Amber protein and ion force field, the TIP4P-D water model, and an in-house force field derived from GAFF.
Simulations:
---
Coarse Grained RBD/ACE2-B0AT1 complex (PDB 6M17)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: RBD ACE2
Model: Files | Source Structure PDBs: 6M17
Coarse graied (CG) mappings are generated from the DSGPM deep learning model.. Chains A,C,E and F of the RBD/ACE2-B0AT1 complex are coarse grained separately. There are 16,16,12 and 12 mappings of different resolutions respectively. These were generated from the original all-atom PDB structure.
Simulations:
---
Human ACE2 ectodomain in aqueous solution (apo open state)
DESRES
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42 | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
DESRES-ANTON-10875753 10 µs simulation trajectory of the human ACE2 ectodomain in aqueous solution
DESRES-ANTON-10875753 10 µs simulation trajectory of the human ACE2 ectodomain, no water or ions
Spike protein in complex with human ACE2
Oostenbrink Lab
University of Natural Resources and Life Sciences, Vienna
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: 6VYB 6M17
Atomistic model of the Spike protein in complex with the human ACE2 receptor, most probale glycosylations are added.
Simulations:
Trajectory of the Spike protein in complex with human ACE2
Structure of SARS coronavirus spike receptor-binding domain complexed with its receptor in aqueous solution
DESRES
Represented Proteins: ACE2 RBD
Model: Files | Source Structure PDBs: 2AJF | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns.
Simulations:
A 10 µs simulation of a SARS-CoV-1 and SARS-CoV-2 chimera-ACE2 complex, no water or ions
Coarse Grained 1r4l Structures (PDB 1R4L ligand removed)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R4L
Coarse graied (CG) mappings prepared with DSGPM deep learning model. There are 20 mappings of different resolutions generated from the original all-atom PDB structure.
Simulations:
---
Native Human Angiotensin Converting Enzyme-Related Carboxypeptidase (ACE2): ISOLDE refined model with glycans
William Glass
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42 | Visualize: 3DMol.js
Takes the ACE2 (Chain A) from the 1r42 refined model, builds all glycans using PyMOL, and refines clashes using ISOLDE. See a complete description of the workflow for building and refining the glycosylated model.
Simulations:
Folding@home simulations of the SARS-CoV-2 spike RBD bound to human ACE2
Human ACE2 ectodomain in aqueous solution (inhibitor-bound closed state)
DESRES
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R4L | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was neutralized and salted with NaCl, with a final concentration of 0.15 M.
Simulations:
DESRES-ANTON-10875754 10 µs simulation trajectory of the human ACE2 ectodomain in aqueous solution
DESRES-ANTON-10875754 10 µs simulation trajectory of the human ACE2 ectodomain, no water or ions
Coarse Grained ACE2 (PDB 1R42 ligands removed)
Geemi Wellawatte, White Lab, University of Rochester
Zhiheng Li, Xu Group, University of Rochester (contributor)
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R42
Coarse graied (CG) mappings are generated from the DSGPM deep learning model. 20 mappings of different resolutions are currently available. The mappings are generated from the original all-atom PDB structure.
Simulations:
---
Chimeric RBD in complex with human ACE2
DESRES
Represented Proteins: ACE2 RBD
Model: Files | Source Structure PDBs: 6VW1 | Visualize: 3DMol.js
The complex model was solvated in a ~140 Å box of 200 mM NaCl and water, and parameterized with the DES-Amber protein and ion force field, the TIP4P-D water model, and an in-house force field derived from GAFF.
Simulations:
---
Docking-based repurposing study of approved drugs against truncated human ACE2 ectodomain (inhibitor-bound closed state)
P. I. Koukos, M. Réau, A. M. J. J Bonvin
Computational Structural Biology group, Bijvoet Centre for Biomolecular Research, Utrecht University
Represented Proteins: ACE2
Model: Files | Source Structure PDBs: 1R4L
Computational docking of the approved subset of Drugbank + active metabolites + investigational compounds of interest against the inhibitor-bound (closed-state) ectodomain of hACE2. The receptor has been truncated by removing residues distant from the binding site to reduce the computational load. 3D conformers were generated from SMILES using OpenEye OMEGA with default settings.
Simulations:
HADDOCK docking of approved Drugbank set against human ACE2 ectodomain
SARS-CoV-2 spike receptor-binding domain bound with ACE2
Negin Forouzesh, Alexey Onufriev
California State University, Los Angeles and Virginia Tech
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Truncated structure of SARS-CoV-2 spike receptor-binding domain bound with the human ACE2 receptor.
Simulations:
SARS-CoV-2 RBD/ACE2-B0AT1 complex in aqueous solution
DESRES
Represented Proteins: ACE2 RBD BoAT1
Model: Files | Source Structure PDBs: 6M17 | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns.
Simulations:
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex, no water or ions
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex in aqueous solution
SWISS-MODEL Spike + ACE2
SWISS-MODEL team, Schwede Group
Swiss Institute of Bioinformatics, Biozentrum, University of Basel
Represented Proteins: spike ACE2
Model: Files | Source Structure PDBs: ---
Full SARS-CoV-2 proteome modeled. Follow the Model link for weekly updated models and PDB structures. All models include global and per-residue quality estimates.
Simulations:
---
SARS-CoV-2 spike receptor-binding domain bound with ACE2: ISOLDE refined model
Tristan Croll
Represented Proteins: spike RBD ACE2
Model: Files | Source Structure PDBs: 6M0J | Visualize: 3DMol.js
Refinement of 6m0j that fixes multiple issues: * Adds/extend GlcNAcs in ACE2 * Fixes backwards His374 (zinc binding site) * Corrects incorrect modeling of both zinc binding sites (originally modeled as disulfides) * Corrects rotamer adjustments and peptide flips See a complete description of the issues remedied by this model.
Simulations:
---
Sodium Dependent Neutral Amnio Acid Transporter (BoAT1)
Blocking SARS-CoV-2 Spike protein binding to human ACE2 receptor
SARS-CoV-2 RBD/ACE2-B0AT1 complex in aqueous solution
DESRES
Represented Proteins: ACE2 RBD BoAT1
Model: Files | Source Structure PDBs: 6M17 | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns.
Simulations:
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex in aqueous solution
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex, no water or ions
Inhibition of formation of the viral fusion core
SARS-CoV-2 RBD/ACE2-B0AT1 complex in aqueous solution
DESRES
Represented Proteins: ACE2 RBD BoAT1
Model: Files | Source Structure PDBs: 6M17 | Visualize: 3DMol.js
The C- and A- peptide termini, including those exposed due to missing loops in the published structural models, are capped with amide and acetyl groups respectively. The system was solvated, neutralized and salted with NaCl, with a final concentration of 0.15 M. The interval between frames is 1.2 ns.
Simulations:
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex in aqueous solution
DESRES-ANTON-10905033 10 µs simulation of the SARS-CoV-2-ACE2 complex, no water or ions
Ab Receptor in Host Cells (FcR)
Furin / PACE
Interleukin-6 (IL-6) receptor
Programmed cell death factor 1
p38 Mitogen-Activated Protein Kinase (MPAK)
Transmembrane Protease Serine 2
No Targets Recorded
TMPRSS2 Homology Model With Benzamidine Overlaid
David Huggins
Tri-Institutional Therapeutics Discovery Institute
Represented Proteins: TMPRSS2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
TMPRSS2 homology model based on the structure of TMPRSS15/Enteropeptidase from PDBID 4DGJ with Benzamidine Overlaid
Simulations:
---
TMPRSS2 Homology Model With Substrate Overlaid
David Huggins
Tri-Institutional Therapeutics Discovery Institute
Represented Proteins: TMPRSS2
Model: Files | Source Structure PDBs: --- | Visualize: 3DMol.js
TMPRSS2 homology model based on the structure of TMPRSS15/Enteropeptidase from PDBID 4DGJ with SARS-CoV-2 substrate peptide overlaid
Simulations:
---
External Model Resources
Structure Models of All Mature Peptides in 2019-NCoV Genome by C-I-TASSER
Description: This page contains 3D structural models and function annotation for all proteins encoded by the genome of 2019-nCoV, also known as SARS-CoV-2, which is the novel coronavirus causing the COVID-19 pneumonia. The structure models are generated by the C-I-TASSER pipeline, which utilizes deep convolutional neural-network based contact-map predictions to guide the I-TASSER fragment assembly simulations. Benchmark and blind CASP tests showed that C-I-TASSER generates models with a higher accuracy than I-TASSER does, especially for the protein targets lack of homologous templates. For multi-domain targets, the C-I-TASSER structure of individual domains are assembled by DEMO into full length structure.
Institution: University of Michigan
Lab: Yang Zhang
Tristan Croll ISOLDE COVID-19 Models
Description: Models refined by Tristan Croll in ISOLDE
Institution: Cambridge Institute for Medical Research
Solvation Maps for COVID19-Related Protein Targets
Description: This is a free online repository for sharing GIST/HSA data of COVID19 related protein targets. Currently, we have th3 GIST/HSA data for 7 targets (6LU7, 6YB7, 6M03, 6Y84, 6W63, 6JYT and 6W4H).
Institution: Lehman College, City University of New York
Lab: Tom Kurtzman
PubChem
Description: PubChem is a public chemistry database at the National Institutes of Health (NIH). Since the launch in 2004, PubChem has become a key chemical information resource for scientists, students, and the general public. Each month our website and programmatic services provide data to several million users worldwide. PubChem mostly contains small molecules, but also larger molecules such as nucleotides, carbohydrates, lipids, peptides, and chemically-modified macromolecules. We collect information on chemical structures, identifiers, chemical and physical properties, biological activities, patents, health, safety, toxicity data, and many others.
Institution: NIH
Structural Biology Task Force GitHub Page
Description: A global public resource for the structures from beta-coronavirus with a focus on SARS-CoV and SARS-CoV-2.
Organization: Coronavirus Structural Biology Task Force