Help
Browser compatibility
| OS | Version | Chrome | Firefox | Microsoft Edge | Safari |
|---|---|---|---|---|---|
| Windows | 10 22H2 | 108.0.5359.99 | n/a | 108.0.1462.46 | n/a |
| macOS | 10.15.7 (Catalina) | 108.0.5359.99 | 107.0, 108.0, 108.0.1 | not tested | 15.6.1 |
| Linux | Ubuntu 20.04 | 108.0.5359 | 107.0 | n/a | n/a |
| Linux | Ubuntu 22.04 | 1:85.0.4183.83 | 108.0 | n/a | n/a |
Minimization
A short molecular dynamics simulation is performed for the top predicted poses in order to relax interactions and removing atomic clashes.
The protocol is performed by the OpenMM framework in the following conditions:
forcefield = ForceField('amber99sb.xml', 'tip3p.xml')
system = forcefield.createSystem(pdb.topology, nonbondedMethod=CutoffNonPeriodic, nonbondedCutoff=0.5*nanometer, constraints=HBonds)
integrator = LangevinIntegrator(300*kelvin, 1/picosecond, 0.002*picoseconds)
simulation = Simulation(pdb.topology, system, integrator)
simulation.context.setPositions(pdb.positions)
simulation.minimizeEnergy()
simulation.step(100)
PDB support
Depending on molecule type, only specific residue or nucleotides might be supported. This also includes specific atom type names. Please find a detailed list of atomic instances compatible with the LightDock Server.
If backbone flexibility is enabled, please make sure all provided residue and nucleotide have no missing backbone atoms. Otherwise ANM protocol might fail.
DFIRE scoring function
DFIRE statistical potential has only support for standard residue types and protonation (hydrogen atoms) are simply removed. A new artificial bead called MMB-BJ is added to the force-field to support fake lipidic membrane boundaries.
| Residue Name | Atom types |
|---|---|
| ALA | N CA C O CB |
| CYS | N CA C O CB SG |
| ASP | N CA C O CB CG OD1 OD2 |
| GLU | N CA C O CB CG CD OE1 OE2 |
| PHE | N CA C O CB CG CD1 CD2 CE1 CE2 CZ |
| GLY | N CA C O |
| HIS | N CA C O CB CG ND1 CD2 CE1 NE2 |
| ILE | N CA C O CB CG1 CG2 CD1 |
| LYS | N CA C O CB CG CD CE NZ |
| LEU | N CA C O CB CG CD1 CD2 |
| MET | N CA C O CB CG SD CE |
| ASN | N CA C O CB CG OD1 ND2 |
| PRO | N CA C O CB CG CD |
| GLN | N CA C O CB CG CD OE1 NE2 |
| ARG | N CA C O CB CG CD NE CZ NH1 NH2 |
| SER | N CA C O CB OG |
| THR | N CA C O CB OG1 CG2 |
| VAL | N CA C O CB CG1 CG2 |
| TRP | N CA C O CB CG CD1 CD2 CE2 NE1 CE3 CZ3 CH2 CZ2 |
| TYR | N CA C O CB CG CD1 CD2 CE1 CE2 CZ OH |
| MMB | BJ |
Protein-Nucleic
Protein-Nucleic scoring function is based on the AMBER94FF types. Protonation is automatically calculated by Reduce software (user input protonation status is removed). Some translation between atom types is required, e.g. OP1 to O1P, and it is automatically performed by the server.
List of supported nucleotides and atom types is detailed below:
| Nucleotide | Atom types |
|---|---|
| DA | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N9, C8, H8, N7, C5, C6, N6, H61, H62, N1, C2, H2, N3, C4, C3', H3', C2', H2'1, H2'2, O3' |
| DC | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N1, C6, H6, C5, H5, C4, N4, H41, H42, N3, C2, O2, C3', H3', C2', H2'1, H2'2, O3' |
| DG | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N9, C8, H8, N7, C5, C6, O6, N1, H1, C2, N2, H21, H22, N3, C4, C3', H3', C2', H2'1, H2'2, O3' |
| DT | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N1, C6, H6, C5, C7, H71, H72, H73, C4, O4, N3, H3, C2, O2, C3', H3', C2', H2'1, H2'2, O3' |
| A | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N9, C8, H8, N7, C5, C6, N6, H61, H62, N1, C2, H2, N3, C4, C3', H3', C2', H2'1, O2', HO'2, O3' |
| C | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N1, C6, H6, C5, H5, C4, N4, H41, H42, N3, C2, O2, C3', H3', C2', H2'1, O2', HO'2, O3' |
| G | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N9, C8, H8, N7, C5, C6, O6, N1, H1, C2, N2, H21, H22, N3, C4, C3', H3', C2', H2'1, O2', HO'2, O3' |
| U | P, O1P, O2P, O5', C5', H5'1, H5'2, C4', H4', O4', C1', H1', N1, C6, H6, C5, H5, C4, O4, N3, H3, C2, O2, C3', H3', C2', H2'1, O2', HO'2, O3' |
Nucleotides should be described in ATOM entries and not in HETATM. LightDock-Server does not support non-standard nucleotides.
Results
LightDock server provides a compressed file containing all relevant information provided and generated by the LightDock simulation, including the top 100 predicted poses by the selected protocol (top/ directory) in the PDB format.
An exhaustive list of files generated during the setup of the simulation can be found here.
All files generated by the LightDock Server and provided in the results page of the job follow the same standard and structure in terms of file format, reproducibility and repeatability of the results as described in the stand-alone version of the LightDock software. Please visit the Tutorials section of the LightDock official website for the most up-to-date information of the LightDock software, including a specific section on how to use this server.
Additionally, users may ask questions directly to the developers via the official Slack space.
Server version
| Software | Version |
|---|---|
| LightDock | 0.9.3.post1 |
| LightDock-Rust | 0.3.0 |
Third-party software
We thank all the third-party software providers used by this server:
| Software | Version |
|---|---|
| PDBFixer | 1.8.1 |
| OpenMM | 7.7 |
| Reduce | reduce.4.9.210817 |
Benchmarking & References
Please cite this reference if you use the LightDock-Server:
The LightDock Server: Artificial Intelligence-powered modeling of macromolecular interactions
Brian Jiménez-García, Jorge Roel-Touris, Didier Barradas-Bautista
Nucleic Acids Research, 2023;, gkad327, doi: https://doi.org/10.1093/nar/gkad327
LightDock has been systematically validated against multiple publicly available datasets and its predictive performance reported in different publications. Please find below a list of all publications and links to public available data repositories:
Publications
Rational Prediction of PROTAC-compatible Protein-Protein Interfaces by Molecular Docking
Paulo C. T. Souza Gilberto P. Pereira, Brian Jiménez-García, Riccardo Pellarin, Guillaume Launay, Sangwook Wu, Juliette Martin
bioRxiv doi: https://doi.org/10.1101/2023.02.16.528819
Integrative Modeling of Membrane-associated Protein Assemblies
Jorge Roel-Touris, Brian Jiménez-García & Alexandre M.J.J. Bonvin
Nat Commun 11, 6210 (2020); doi: https://doi.org/10.1038/s41467-020-20076-5
LightDock goes information-driven
Jorge Roel-Touris, Alexandre M.J.J. Bonvin and Brian Jiménez-García
Bioinformatics, Volume 36, Issue 3, 1 February 2020, Pages 950–952, doi: https://doi.org/10.1093/bioinformatics/btz642
LightDock: a new multi-scale approach to protein–protein docking
Brian Jiménez-García, Jorge Roel-Touris, Miguel Romero-Durana, Miquel Vidal, Daniel Jiménez-González and Juan Fernández-Recio
Bioinformatics, Volume 34, Issue 1, 1 January 2018, Pages 49–55, doi: https://doi.org/10.1093/bioinformatics/btx555
Public available data repositories
| Description | Repository |
|---|---|
| Protein-protein Docking Benchmark 5 LightDock performance evaluated by DockQ | https://github.com/lightdock/bm5 |
| The performance of the residue restraint driven protocol evaluated on the new 55 cases of the protein-protein Docking Benchmark 5 | https://github.com/lightdock/lightdock_bm5 |
| Integrative Modeling of Membrane-associated Protein Assemblies: Dataset | https://github.com/lightdock/membrane_docking |
| Restraints how-to and examples | https://github.com/lightdock/test_restraints |
LightDock is a FOSS project, all data and code is public and available at https://github.com/lightdock.