Molecular Structure Checking using BioExcel Building Blocks (biobb)

This tutorial aims to illustrate the process of checking a molecular structure before using it as an input for a Molecular Dynamics simulation. The workflow uses the BioExcel Building Blocks library (biobb). The particular structure used is the crystal structure of human Adenylate Kinase 1A (AK1A), in complex with the AP5A inhibitor (PDB code 1Z83).

Structure checking is a key step before setting up a protein system for simulations. A number of common issues found in structures at Protein Data Bank may compromise the success of the simulation, or may suggest that longer equilibration procedures are necessary.

The workflow shows how to:

Run basic manipulations on structures (selection of models, chains, alternative locations
Detect and fix amide assignments and wrong chiralities
Detect and fix protein backbone issues (missing fragments, and atoms, capping)
Detect and fix missing side-chain atoms
Add hydrogen atoms according to several criteria
Detect and classify atomic clashes
Detect possible disulfide bonds (SS)

NOTE:

Some of the workflow steps work with AMBER force-field nomenclature (e.g. CYX, HID, HIE), and thus the output structure might be slightly modified to use it with different force-fields.

An implementation of this workflow in a web-based Graphical User Interface (GUI) can be found in the https://mmb.irbbarcelona.org/biobb-wfs/ server (see https://mmb.irbbarcelona.org/biobb-wfs/help/create/structure#check).

Settings

Biobb modules used

biobb_io: Tools to fetch biomolecular data from public databases.
biobb_structure_utils: Tools to modify or extract information from a PDB structure.
biobb_model: Tools to check and model 3D structures, create mutations or reconstruct missing atoms.
biobb_amber: Tools to setup and simulate atomistic MD simulations using AMBER MD package.
biobb_chemistry: Tools to perform chemoinformatics on molecular structures.

Auxiliar libraries used

modeller: Software used for homology or comparative modeling of protein three-dimensional structures.
jupyter: Free software, open standards, and web services for interactive computing across all programming languages.
nglview: Jupyter/IPython widget to interactively view molecular structures and trajectories in notebooks.
plotly: Python interactive graphing library integrated in Jupyter notebooks.

Conda Installation and Launch

git clone https://github.com/bioexcel/biobb_wf_structure_checking.git
cd biobb_wf_structure_checking
conda env create -f conda_env/environment.yml
conda activate biobb_wf_structure_checking
jupyter-notebook biobb_wf_structure_checking/notebooks/biobb_wf_structure_checking.ipynb

Bioexcel2 logo

Input parameters

Input parameters needed:

pdbCode: PDB code of the protein structure (e.g. 6M0J)

In [ ]:

import nglview
import ipywidgets

pdbCode = "1z83" # Should be replaced to check different structures

Fetching PDB structure

Downloading PDB structure with the protein molecule from the RCSB PDB database.
Alternatively, a PDB file can be used as starting structure.

Building Blocks used:

Pdb from biobb_io.api.pdb

In [ ]:

# Downloading desired PDB file 
# Import module
from biobb_io.api.pdb import pdb

# Create properties dict and inputs/outputs
downloaded_pdb = pdbCode+'.pdb'
prop = {
    'pdb_code': pdbCode,
    'filter': ['ATOM', 'HETATM'],
}

#Create and launch bb
pdb(output_pdb_path=downloaded_pdb,
    properties=prop)

Visualizing 3D structure

Visualizing the downloaded/given PDB structure using NGL:

In [ ]:

# Show original structure
view = nglview.show_structure_file(downloaded_pdb)
view.add_representation(repr_type='spacefill', selection='ion')
view.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view._remote_call('setSize', target='Widget', args=['','400px'])
view

PDB structure checking

The first step of the workflow is a complete checking of the molecular structure. The BioBB utility structure_check parses the whole PDB file, runs a complete structure checking, saves useful information about the structure and looks for possible issues. The block is printing a report of the information extracted from the input structure.

Building Blocks used:

structure_check from biobb_structure_utils.utils.structure_check

In [ ]:

from biobb_structure_utils.utils.structure_check import structure_check

report = pdbCode + ".report.json"

structure_check(
    input_structure_path=downloaded_pdb,
    output_summary_path=report
)

Structure checking, step by step

The workflow will go through the report, exploring all the different features analysed. The following cell is showing the full content of the report.

In [ ]:

import json
with open(report, 'r') as f:
  data = json.load(f)
print(json.dumps(data, indent=2))

Models

Presence of models in the structure. MD simulations require a single structure, although some structures (e.g. biounits) may be defined as a series of models, in such case all of them are usually required.

Use BioBB extract_model to select the desired models.

extract_model from biobb_structure_utils.utils.extract_model

In [ ]:

print(json.dumps(data['models'], indent=2))

In [ ]:

from biobb_structure_utils.utils.extract_model import extract_model

pdb_models = pdbCode + ".models.pdb"

prop = {
    'models': [ 1 ]
}

extract_model(
    input_structure_path=downloaded_pdb,
    output_structure_path=pdb_models,
    properties=prop
)

In [ ]:

# Show modified structure
view = nglview.show_structure_file(pdb_models)
view.add_representation(repr_type='spacefill', selection='ion')
view.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view._remote_call('setSize', target='Widget', args=['','400px'])
view

Chains

Presence of chains in the structure. MD simulations are usually performed with complete structures. However input structure may contain several copies of the system, or contain additional chains like peptides or nucleic acids that may be removed.

Use BioBB extract_chain to select the desired chains.

extract_chain from biobb_structure_utils.utils.extract_chain

In [ ]:

print(json.dumps(data['chains'], indent=2))

In [ ]:

from biobb_structure_utils.utils.extract_chain import extract_chain

pdb_chains = pdbCode + ".chains.pdb"

prop = {
    'chains': [ 'A' ]
}

extract_chain(
    input_structure_path=pdb_models,
    output_structure_path=pdb_chains,
    properties=prop
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_models)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1._remote_call('setSize', target='Widget', args=['','600px'])
view2 = nglview.show_structure_file(pdb_chains)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2._remote_call('setSize', target='Widget', args=['','600px'])
ipywidgets.HBox([view1, view2])

Alternative Locations

Presence of residues with alternative locations. Atoms with alternative coordinates and their occupancy are reported. MD simulations requires a single position for each atom.

Use BioBB fix_altloc to select the desired alternative locations.

fix_altlocs from biobb_model.model.fix_altlocs

In [ ]:

print(json.dumps(data['altloc'], indent=2))

In [ ]:

from biobb_model.model.fix_altlocs import fix_altlocs

pdb_altloc = pdbCode + ".altloc.pdb"

prop = {
    'altlocs': ['A45:A', 'A67:A', 'A85:A'] 
}

fix_altlocs(
    input_pdb_path=pdb_chains,
    output_pdb_path=pdb_altloc,
    properties=prop
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_chains)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1.add_representation(repr_type='ball+stick', radius='0.4', selection='45%A 67%A 85%A')
view1.add_representation(repr_type='ball+stick', radius='0.2', selection='45%B 67%B 85%B')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_altloc)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2.add_representation(repr_type='ball+stick', radius='0.4', selection='45 67 85')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Disulfide Bridges

Presence of disulfide bridges (-S-S- bonds) based on distance criteria. Disulfide bonds are crucial for the structure stability, and MD simulations require those bonds to be correctly set. Most of the MD packages come with tools able to automatically identify -S-S- bonds and include them in the system topology parameters, but some of them need the residues involved in the bond to be explicitly marked (e.g. AMBER CYX residues).

Use BioBB fix_ss_bonds to mark the disulfide bridges.

fix_ssbonds from biobb_model.model.fix_ssbonds

In [ ]:

print(json.dumps(data['getss'], indent=2))

In [ ]:

from biobb_model.model.fix_ssbonds import fix_ssbonds

pdb_ssbonds = pdbCode + ".ssbonds.pdb"

fix_ssbonds(
    input_pdb_path=pdb_altloc,
    output_pdb_path=pdb_ssbonds
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_altloc)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1.add_representation(repr_type='ball+stick', radius='0.4', selection='CYS')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_ssbonds)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2.add_representation(repr_type='ball+stick', radius='0.2', selection='CYS')
view2.add_representation(repr_type='ball+stick', radius='0.4', selection='CYX')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Metal Ions

Presence of heteroatoms being metal ions. Only structural metal ions should be kept in MD simulations, as they require special force field parameters. In some cases, the atomic coordination between the protein atoms and the metal ions should be also specified.

Use BioBB remove_molecules to remove the undesired metal ions.

remove_molecules from biobb_structure-utils.utils.remove_molecules

In [ ]:

print(json.dumps(data['metals'], indent=2))

In [ ]:

from biobb_structure_utils.utils.remove_molecules import remove_molecules

pdb_metals = pdbCode + ".metals.pdb"

prop = {
    'molecules': [
        {
            'name': 'ZN'
        }
    ]
}
remove_molecules(input_structure_path=pdb_ssbonds,
                 output_molecules_path=pdb_metals,
                 properties=prop)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_chains)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1.add_representation(repr_type='ball+stick', radius='0.4', selection='CYS')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_metals)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2.add_representation(repr_type='ball+stick', radius='0.2', selection='CYS')
view2.add_representation(repr_type='ball+stick', radius='0.4', selection='CYX')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Ligands

Presence of heteroatoms being ligands. Only important ligands should be kept in MD simulations, as they require special force field parameters.

Use BioBB remove_molecules to remove the undesired ligands.

remove_molecules from biobb_structure-utils.utils.remove_molecules

In [ ]:

print(json.dumps(data['ligands'], indent=2))

In [ ]:

from biobb_structure_utils.utils.remove_molecules import remove_molecules

pdb_ligands = pdbCode + ".ligands.pdb"

prop = {
    'molecules': [
        {
            'name': 'SO4',
        },
        {
            'name' : 'AP5'
        }
    ]
}
remove_molecules(input_structure_path=pdb_metals,
                 output_molecules_path=pdb_ligands,
                 properties=prop)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_metals)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_ligands)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Hydrogen atoms

Presence of hydrogen atoms. MD setup can be done with the original hydrogen atoms, however to prevent possible problems coming from non standard labelling, removing them is safer.

Use BioBB reduce_remove_hydrogens to remove the undesired hydrogen atoms.

reduce_remove_hydrogens from biobb_chemistry.ambertools.reduce_remove_hydrogens

In [ ]:

print(json.dumps(data['stats']['stats'], indent=2))

In [ ]:

from biobb_chemistry.ambertools.reduce_remove_hydrogens import reduce_remove_hydrogens

pdb_hydrogens = pdbCode + ".hydrogens.pdb"

reduce_remove_hydrogens(
    input_path=pdb_ligands,
    output_path=pdb_hydrogens
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_ligands)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1.add_representation(repr_type='spacefill', selection='hydrogen')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_hydrogens)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2.add_representation(repr_type='spacefill', selection='hydrogen')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Water molecules

Presence of water molecules. Crystallographic water molecules may be relevant for keeping the structure stable, however in most cases only some of them are required. These can be later added using other methods.

Use BioBB remove_pdb_water to remove the undesired water molecules.

remove_pdb_water from biobb_structure-utils.utils.remove_pdb_water

In [ ]:

print(json.dumps(data['stats']['stats'], indent=2))

In [ ]:

from biobb_structure_utils.utils.remove_pdb_water import remove_pdb_water

pdb_water = pdbCode + ".water.pdb"

remove_pdb_water(
    input_pdb_path=pdb_hydrogens,
    output_pdb_path=pdb_water
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_hydrogens)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.5', selection='water')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_water)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.5', selection='water')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Amide groups

Presence of incorrect amide groups. Amide terminal atoms in Asparagines and Glutamines residues can be labelled incorrectly, with swapped Nitrogen and Oxygen atoms from the amide group.

Use BioBB fix_amides to fix the incorrectly assigned amide groups.

fix_amides from biobb_model.model.fix_amides

In [ ]:

print(json.dumps(data['amide'], indent=2))

In [ ]:

from biobb_model.model.fix_amides import fix_amides

pdb_amides = pdbCode + ".amides.pdb"

fix_amides(
    input_pdb_path=pdb_water,
    output_pdb_path=pdb_amides
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_water)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1.add_representation(repr_type='ball+stick', radius='0.4', selection='amide')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_amides)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2.add_representation(repr_type='ball+stick', radius='0.4', selection='amide')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Chirality

Presence of incorrect chiralities. Side chains of Threonine and Isoleucine residues are chiral.
Incorrect atom labelling can lead to the wrong chirality.

Use BioBB fix_chirality to fix the incorrectly assigned chirality.

fix_chirality from biobb_model.model.fix_chirality

In [ ]:

print(json.dumps(data['chiral'], indent=2))

In [ ]:

from biobb_model.model.fix_chirality import fix_chirality

pdb_chiral = pdbCode + ".chiral.pdb"

fix_chirality(
    input_pdb_path=pdb_amides,
    output_pdb_path=pdb_chiral
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_amides)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_chiral)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Side Chains

Presence of missing protein side chain atoms. MD programs only work with complete residues; some of the MD packages come with tools able to model missing side chain atoms, but not all of them.

Use BioBB fix_side_chain to model the missing side chain atoms.

fix_side_chain from biobb_model.model.fix_side_chain

In [ ]:

print(json.dumps(data['fixside'], indent=2))

In [ ]:

from biobb_model.model.fix_side_chain import fix_side_chain

pdb_side_chains = pdbCode + ".sidechains.pdb"

fix_side_chain(
    input_pdb_path=pdb_chiral,
    output_pdb_path=pdb_side_chains
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_chiral)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1.add_representation(repr_type='ball+stick', radius='0.4', selection='7 56 63 123 127 143 155')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_side_chains)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2.add_representation(repr_type='ball+stick', radius='0.4', selection='7 56 63 123 127 143 155')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Backbone

Presence of missing backbone atoms. PDB files can have missing backbone atoms, usually placed in flexible regions of the molecule. MD programs work with amino acid libraries that need all atoms to be present. Some MD packages tools are able to model side chain missing atoms, but they are rarely capable of modeling backbone atoms.

Use BioBB fix_backbone to model the missing backbone atoms.

fix_backbone from biobb_model.model.fix_backbone

NOTE: This building block uses Modeller to rebuilt missing backbone atoms. Modeller requires a license that can be easily obtained in the Modeller registration page. This license in form of a keyword needs to be written in a specific file in order for Modeller to work. This file is shown in the next cell. Alternatively, the keyword can be inserted using the modeller_key property of the building block.

An additional building block is used to extract the canonical FASTA sequence for the protein, to be used in Modeler to model the missing backbone region:

canonical_fasta from biobb_io.api.canonical_fasta

In [ ]:

print(json.dumps(data['backbone'], indent=2))

In [ ]:

from biobb_io.api.canonical_fasta import canonical_fasta

pdb_fasta = pdbCode + ".fasta"

prop = {
    'pdb_code': pdbCode,
    'api_id': 'pdb'
}

canonical_fasta(
    output_fasta_path=pdb_fasta,
    properties=prop
)

In [ ]:

# Add Modeller key in the config file before running the next cell
# Alternatively, add the Modeller key in the "modeller_key" property of the fix_backbone building block
import os
conda_env_path = os.environ['CONDA_PREFIX']
modeller_key_path = conda_env_path + "/lib/modeller-10.3/modlib/modeller/config.py"
print("WARNING: Edit this file and add your Modeller KEY:\n " + modeller_key_path)

In [ ]:

from biobb_model.model.fix_backbone import fix_backbone

pdb_backbone = pdbCode + ".backbone.pdb"

prop = {
    
}

fix_backbone(
    input_pdb_path=pdb_side_chains,
    input_fasta_canonical_sequence_path=pdb_fasta,
    output_pdb_path=pdb_backbone,
    properties=prop
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_side_chains)
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_backbone)
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Atomic Clashes

Presence of atomic clashes. Atoms that are too close in space can have a problem of energetic repulsion. Most clashes come from over-compactation of crystal structures and are naturally corrected on system setup or MD equilibration, but may lead to a significant distortion of the structure. Clashes are detected based on distance criteria and are classified in different groups, depending on the atom types involved:

Severe: Atoms too close, usually indicating superimposed structures or badly modelled regions. Should be fixed.
Apolar: Vdw collisions. Usually fixed during the simulation.
Polar and ionic: Usually indicate wrong side chain conformations. Usually fixed during the simulation

Use the BioBB_amber module to energetically minimize the structure and fix atomic clashes.

leap_gen_top from biobb_amber.leap.leap_gen_top
sander_mdrun from biobb_amber.sander.sander_mdrun
amber_to_pdb from biobb_amber.ambpdb.amber_to_pdb

In [ ]:

print(json.dumps(data['clashes'], indent=2))

In [ ]:

from biobb_amber.leap.leap_gen_top import leap_gen_top

pdb_amber = pdbCode + ".amber.pdb"
top_amber = pdbCode + ".amber.top"
crd_amber = pdbCode + ".amber.crd"

prop = {
    'forcefield': ['protein.ff14SB']
}

leap_gen_top(
    input_pdb_path=pdb_backbone,
    output_pdb_path=pdb_amber,
    output_top_path=top_amber,
    output_crd_path=crd_amber,
    properties=prop
)

In [ ]:

from biobb_amber.sander.sander_mdrun import sander_mdrun

trj_amber = pdbCode + ".amber.crd"
rst_amber = pdbCode + ".amber.rst"
log_amber = pdbCode + ".amber.log"

prop = {
    'simulation_type' : 'minimization',
    'mdin' : {
        'ntb' : 0,        # Periodic Boundary. No periodicity is applied and PME (Particle Mesh Ewald) is off.
        'cut' : 12,       # Nonbonded cutoff, in Angstroms.
        'maxcyc' : 500,   # Maximum number of cycles of minimization.
        'ncyc' : 50       # Minimization will be switched from steepest descent to conjugate gradient after ncyc cycles.
    }
}

sander_mdrun(
     input_top_path=top_amber,
     input_crd_path=crd_amber,
     output_traj_path=trj_amber,
     output_rst_path=rst_amber,
     output_log_path=log_amber,
     properties=prop
)

In [ ]:

from biobb_amber.ambpdb.amber_to_pdb import amber_to_pdb

pdb_amber_min = pdbCode + ".amber-min.pdb"

amber_to_pdb(
      input_top_path=top_amber,
      input_crd_path=rst_amber,
      output_pdb_path=pdb_amber_min
)

In [ ]:

# Show modified structure
view1 = nglview.show_structure_file(pdb_backbone)
view1.add_representation(repr_type='ball+stick', selection='all')
view1.add_representation(repr_type='spacefill', selection='ion')
view1.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view1._remote_call('setSize', target='Widget', args=['','450px'])
view2 = nglview.show_structure_file(pdb_amber_min)
view2.clear_representations()
view2.add_representation(repr_type='cartoon', selection='all')
view2.add_representation(repr_type='ball+stick', selection='all')
view2.add_representation(repr_type='spacefill', selection='ion')
view2.add_representation(repr_type='spacefill', radius='0.3', selection='water')
view2._remote_call('setSize', target='Widget', args=['','450px'])
ipywidgets.HBox([view1, view2])

Fix PDB

Information included in the original structure is sometimes lost during the preparation process. Typical examples are the residue numeration and the structure chains, that are changed or even removed by some MD tools (as happened with the chains in the previous energy minimization process). The fix_pdb building block renumerates the residues in a PDB structure according to reference sequences from UniProt and assigns chain ids accordingly.

Use the BioBB_model module to fix the structure, renumerating the residues and assigning chain ids.

fix_pdb from biobb_model.model.fix_pdb

In [ ]:

from biobb_model.model.fix_pdb import fix_pdb

final_pdb = pdbCode + ".final.pdb"

fix_pdb(
    input_pdb_path=pdb_amber_min,
    output_pdb_path=final_pdb
)

Final Check

Checking the final structure to analyse the possible issues present in the modified structure in comparison with the issues we found with the original structure at the beginning of the workflow.

Building Blocks used:

structure_check from biobb_structure_utils.utils.structure_check

In [ ]:

from biobb_structure_utils.utils.structure_check import structure_check

report_final = pdbCode + ".report_final.json"

structure_check(
    input_structure_path=final_pdb,
    output_summary_path=report_final
)

In [ ]:

import json
with open(report_final, 'r') as f:
  data_final = json.load(f)
print(json.dumps(data_final, indent=2))

Printing the possible issues identified by the structure_check building block in the original structure (left column) against the checked structure (right column). The main differences are the elimination of the alternative locations, metal ions and ligands. Most of the clashes have also disappeared thanks to the short energetic minimization.

The structure should be now ready to start the MD setup process.

In [ ]:

from ipywidgets import Layout

output1 = ipywidgets.widgets.Output(layout=Layout(width='50%'))
with output1:
    display(print(json.dumps(data, indent=2)))
output2 = ipywidgets.widgets.Output(layout=Layout(width='50%'))
with output2:
    display(print(json.dumps(data_final, indent=2)))

two_columns = ipywidgets.widgets.HBox([output1, output2])

display(two_columns)

Questions & Comments

Questions, issues, suggestions and comments are really welcome!

GitHub issues:
- https://github.com/bioexcel/biobb
BioExcel forum:
- https://ask.bioexcel.eu/c/BioExcel-Building-Blocks-library

Molecular Structure Checking tutorial

Molecular Structure Checking using BioExcel Building Blocks (biobb)

Settings

Biobb modules used

Auxiliar libraries used

Conda Installation and Launch

Pipeline steps

Input parameters

Fetching PDB structure

Visualizing 3D structure

PDB structure checking

Structure checking, step by step

Models

Chains

Alternative Locations

Disulfide Bridges

Metal Ions

Ligands

Hydrogen atoms

Water molecules

Amide groups

Chirality

Side Chains

Backbone

Atomic Clashes

Fix PDB

Final Check

Questions & Comments