Molecular Docking Tutorial
Presented By:Diksha Marwaha
Course Instructor:[Link] Ray
Course -Biophysics(Bio215)
What is Molecular Docking?
It is a computational method that predicts orientation of
ligand when binds in an active site of target receptor to
form stable complexes virtually.
Use-Reliable prediction of docking result is useful in
structure-based drug designing.
Goal-To achieve optimized confirmation and relative
orientation between ligand and receptor so that free energy
of overall system is minimized.
It consist of two steps:
1. Docking- generation of receptor-ligand poses
within binding cavity.
2. Scoring- Energetic evaluation of poses to find best
scoring pose on the basis of minimum binding
affinity or stability.
Components of Docking
● Target Receptor-macromolecules like protein, DNA, or
RNA
● Ligand-small molecule(atom or ion) or drug candidates that
bind to receptor.
● Active Site- it is specific region on the receptor where
ligand binds and results in conformation change of receptor.
Classification of Docking
On the basis of :
● flexibility of the interacting molecules.
● Type of Interacting molecules.
● Active site
Types of Docking-on basis of flexibility
● Rigid Docking(Lock and Key)-The shape of both ligand and
receptor is rigid and ligand acts as key to rotate and move to see
if it fits in pocket of receptor as it is or [Link] for initial
screening of thousands of molecules.
● Flexible Docking(Induced Fit)-it follows induced fit model
where two molecules actually change shape dynamically(twist
and bend).it has higher accuracy than rigid but much slower.
Types of Docking-Type of Interacting Molecules
Protein-Ligand Docking-To predict how Protein-Protein Docking- Modeling the Protein-Nucleic Acid Docking-Modeling
small molecule (drugs) bind to protein. interactions between two large proteins. the interactions between proteins and DNA
or RNA.
AutoDock Vina, AutoDock 4, GLIDE HADDOCK, ZDOCK
rDock, RLDOCK, XDock
Types of Docking-Active Site
Blind Docking-when active site on receptor is unknown so entire protein surface is scanned for potential binding pockets.
Active-site Docking- when active site is already known.
Inverse Docking- Dock single ligand against many different proteins to identify off-target effects.
Difference between Single Ligand Docking and Virtual Screening
Single Ligand Docking Virtual Screening
Predicts optimal binding pose and affinity of a ligand Identify high-affinity ligand for target by searching high
within target receptor active site chemical space.
Already have known reference ligand Ligand is unknown
Use case is Validation Use case is Drug Discovery
Output is ranked ligand list Output is detailed Interaction
Molecular Docking -Workflow
Requirements and Installations
1. MGL Tools([Link] )
Install version 1.5.7 for your OS(Windows, mac or linux)
2. OpenBabel ([Link] ) for
windows and
3. AutoDock Vina
4. Chimera
5. Protein receptor from RCSB PDB database
6. Ligands from PubChem database
File Formats
PDB:Protein Data Bank format file
SDF:Structure Data file
PDBQT:Protein Data Bank, Partial charge(Q), & Atom type(T) format
GPF:Grid Parameter File
GLG:Grid Log File
DPF:Docking Parameter File
DLG:Docking Log File
Step1: Retrieval of Target protein and Ligand
1. Download
target protein
receptor from To get multiple ligands
PDB database for target receptor,
2. [Link]
[Link]/ download from
PDBID:3DTC PubChem in 3D SDF
3. Download format.
known ligand
(C27 H24 N2 O3)
from PDB in
3D SDF
format or from
PubChem.
3DTC-Crystal structure
of mixed-lineage kinase
MLK1 complexed with
compound 16 consist of
chain A
Step2: Ligand and Receptor formats conversion via OpenBabel
Convert both ligand n SDF and protein in PDB into PDBQT formats.
Open cmd and give command :
obabel [Link] -O [Link]
obabel [Link] -O [Link]
Alternatively, Obabel GUI can also be used.
Step3: Target protein Preparation
[Link] AutoDock tools
2. Set the preferred working directory
3. File>read molecule>[Link]
4. To remove water molecules and heteroatoms
(inhibitors or ligands):
Select chain A>Edit>Delete water
Select chain A hetero atom>Edit>delete selected
atoms
5. To prepare missing atoms:
Edit>misc>check for missing atoms
Edit>misc>Repair missing atoms>Save as 2
sets
[Link] Hydrogen atoms:
Edit>Hydrogen>polar only
7. Add Charge:
Edit>charges>Add kollman charges
This means now the
8. Add Charge: protein is completed
charged.
Edit>charges>check total on residues
[Link] receptor as PDBqt
Grid >
Macromolecule>choose>[Link]
Step4: Ligand Preparation
[Link]>[Link]
[Link] ligand is already setup with charges.
[Link]>Torsion tree> Detect root.
[Link]>Torsion tree>set no of Torsions
There are 5 no of active torsions in the ligand
as shown by white arrow in figure.
No of torsions means that no of rotatable
bonds and that no of conformations will be
generated or take more time for docking.
Step5: Active site
1. Get the Active site
residues information from
reported literature.
2. Reported residues were
GLU221,
ALA223,ARG230,
SER272 of receptor
MLK1 for the ligand.
3. Select>Select from
String>Residue>Add>Dis
miss
4. Find X,Y,Z coordinates of
active site of crystal
structure([Link])
Step5: Grid Generation
1. Copy and paste the last atom coordinates of
active site amino acids, sum up each coordinate
and divide it by number of residues
2. Grid>Macromolecule>choose [Link]>ok
3. Grid>Grid box.
4. Enter these X, Y, Z coordinates in the “centre
Grid Box”
5. It will then exactly find location of Active site and
covered by grid box,
6. File>Save as current.
7. Create confi[Link] file and save in same directory.
Step6: Docking via vina
1. Open cmd in windows and run : receptor = [Link]
ligand = [Link]
vina --config confi[Link] --out
center_x = 60
[Link]
center_y = 60
center_z = 60
2. Best affinity = -5.082 kcal/mol
size_x = 25.70975
size_y = 38.33725
size_z = 29.99825
exhaustiveness = 8
num_modes = 10
confi[Link]
Step7: Visualization of docked complex via Chimera
Note :- Model with lowest binding affinity
represents the most stable state. Because it
estimates strength of binding between ligand and
target protein represented as gibbs free energy in
Kcal/mol. More negative the binding affinity,
signifies more favourable binding.
In order to visualize all the conformations-
Open Chimera > Open receptor PDB > tools >
structure binding analysis > ViewDock > open
[Link]