Integrative Modeling of Membrane-associated Protein Assemblies: Dataset

1. Introduction

This is the repository containing the dataset for the Integrative Modeling of Membrane-associated Protein Assemblies manuscript.

2. Data organization

In top folder:

• pdbs.list: list of all the PDB codes of the benchmark.
• antibodies.list: list of all the PDB codes of the benchmark which are antibody complexes (category A in figure).
• barrel.list: list of all the PDB codes of the benchmark which structure is beta-barrel (category B in figure).
• helical.list: list of all the PDB codes of the benchmark which structure is helical (category C in figure).

2.1. Docking

Docking protocols can be found in the docking folder. For each of the two protocols (LightDock and ZDock), data for the different scenarios is provided:

2.1.1. LightDock

Folder docking/lightdock containts data for the different six scenarios: average, blind, filtered, membrane, minimum and restraints.

Each of the scenarios contains for the 18 complexes of the benchmark the following data:

• lgd_clustered_rank.list: Ranking of the top clustered structures generated by LightDock. For each line, name of the PDB complex file and values for fnat, i-RMSD, l-RMSD and LightDock score are provided.
• top100.tgz: Top 100 predicted models generated by LightDock.
• setup.json: LightDock generated setup file which contains information about all the parameters used for the setup of the simulation.
• receptor_membrane.pdb or receptor.pdb: Starting receptor PDB structure. If file contains membrane beads, name contains the word membrane.
• ligand.pdb: Starting ligand PDB structure
• lightdock_* files: LightDock parsed and processed PDB structures.
• restraints.list: residue restraints in the LightDock-ready format.

2.1.2. ZDock

Folder docking/zdock containts data for the different four scenarios: average, blind, maximum and minimum.

Each of the scenarios contains for the 18 complexes of the benchmark the following data:

• zdock_rank.list: Ranking of the top structures generated by ZDock. For each line, name of the PDB complex file and values for fnat, i-RMSD and l-RMSD are provided.
• zdock.out: Result of the ZDock simulation.
• top100.tgz: Top 100 predicted models generated by ZDock.
• receptor.pdb and ligand.pdb: PDB files for the starting structures (receptor and ligand respectively).
• *.block: list of residues blocked for the scenario. In case of antibody-antigen complex, a cdr.block file is also provided with the list of blocked residues corresponding to the CDR predicted loops.
• receptor_rotated.pdb and ligand_rotated.pdb: PDB files for the starting structures after being randomly rotated to avoid simulation artifacts (receptor and ligand respectively).
• receptor_blocked.pdb and ligand_blocked.pdb: PDB files with blocked residues according to the ZDock protocol (receptor and ligand respectively).
• receptor_surface.pdb and ligand_surface.pdb: PDB files with marked surface residues according to the ZDock protocol (receptor and ligand respectively).

2.2. Refinement

Refinement with HADDOCK-CG protocol can be found at refinement/haddock. Results of the refinement protocol for two scenarios are provided: blind and membrane. Each of these scenarios contains a folder for each of the complexes of the dataset. For each complex, two files are provided:

• haddock_rank.list: for each of the diffeerent models refined, values for fnat, i-RMSD, l-RMSD and scoring is provided. For example:

1k4c_80w.pdb   0.0   20.182   50.751   -217.43964

• models.tgz: compressed tar archive which contains the top 100 models refined by the HADDOCK-CG protocol.

In addition, an example of run.param and run.cns for performing HADDOCK refinement is provided in refinement/example.