,

Faculty of Biomedical Engineering

BME310
2023-2024

Experiment #3:
Molecular Docking with Modified Aspirin

Name Surname: Can Mungan

Student ID:20160612031

1. Introduction
 The continuous advancement of bioinformatics relies on the rapid development of
innovative algorithms to meet emerging challenges. Bioinformatics faces the complex task of
managing large datasets containing screening results and molecular structures, which are
essential for future bioinformatics tools. Screening laboratories generate more than 100,000 data
points daily, and compound repositories already exceed one million compounds, often referred to
as the 'magic threshold.' In addition, synthetic combinatorial libraries with over 10,000 chemicals
contribute to virtual combinatorial libraries housing more than 10 million molecules.
Future bioinformatics algorithms must be capable of applying molecular modeling
techniques to these extensive compound libraries, allowing the identification of common patterns
in structural, pharmacological, or property spaces. This involves calculating various molecular
and structural descriptors, as well as physico-chemical characteristics such as lipophilicity and
drug-likeness for large compound sets. Data-mining techniques are then used to establish
connections between descriptor similarity and chemical or biological similarity.
Recent advancements in drug discovery heavily rely on automated methodologies,
necessitating the swift adaptation of computational methods. To address these evolving
challenges, leading commercial molecular modeling software developers are expanding their
offerings to include bioinformatics components. However, these commercial solutions often rely
on complex data management systems and proprietary computing techniques.
In contrast, Marvin offers a straightforward and accessible software package concept. It
serves as a prototype platform that eases the rapid development of novel bioinformatics
algorithms by integrating modules from existing molecular modeling software and enabling the
creation of new programs. Moreover, Marvin simplifies the testing and setup processes,
streamlining the application of new techniques to existing datasets. Due to its user-friendly
approach, it can be implemented on various computing platforms. Marvin's open interface allows
for the concurrent utilization of multiple modules and computational techniques, thanks to its
modular design. This means that most new bioinformatics algorithms can be integrated by
incorporating pre-existing software packages, requiring minimal effort to create a new algorithm
from the ground up.
In the context of this study, the Aspirin molecule was specifically tailored to function as a
ligand and was modified with new molecules. This customized ligand will be used in docking
analyses with 6MQF and 6XLI molecules, which are known to interact with Aspirin. The study
will employ the PyMOL program to further explore these interactions and their implications.
2. Materials & Methods
Tools:
-AutoDock Vina
-AutoDock Tools
-Protein in PDB (Protein Data Bank) format
-Ligand
-PyMOL
 We started by examining the characteristics of the Aspirin molecule on PubChem.
 Then, we downloaded the 2D structure of Aspirin in SDF format.
 To work with the molecule, we accessed Marvin JS through the RCSB website (by
clicking a link, going to the Search section, and using the Chemical Sketch Tool).
 Using Marvin JS, we opened the Aspirin molecule.
 We made modifications to the Aspirin molecule by adding other elements, all within
Marvin JS.
 After the modifications, we converted the newly altered Aspirin into a 3D format
using Marvin JS.
 We then downloaded this modified Aspirin molecule.
 Next, we delved into investigating proteins associated with Aspirin on PubChem.
 These proteins were accessed and opened in the PDB format.
 We specifically selected and downloaded the 6MQF protein structure.
 To prepare for molecular docking, we utilized AutoDockTools. We loaded the "vina"
file as the ligand and the protein structure from 6mqf.pdb.
 Hydrogen atoms were added to the protein structure using the "Edit → Hydrogens →
Add → Polar Only" command.
 We defined the search space for docking by setting grid parameters (Grid →
Macromolecule → Choose Protein → Select Molecule).
 Configurations for the grid were specified, which entailed setting the dimensions to
22 points in X, 24 points in Y, and 28 points in Z, with a spacing of 1.000 Ångstrom.
 The center coordinates for the search space were established at X: 11, Y: 90.5, Z:
57.5.
 We prepared the ligand by loading its structure from the "vina" file and hid the
protein for better visualization.
 Rotatable bonds in the ligand were defined.
 The prepared ligand structure was saved in .pdbqt format.
 We created a configuration file called "conf," which outlined the receptor, ligand,
center coordinates, and size.
 To initiate the molecular docking process, we ran Vina from the command prompt,
navigating to the "vina" directory.
 The Vina executable was utilized with the configuration file, which generated output
binding modes stored in the ligand_out file.
 Finally, we examined the results by using PyMOL, where we visualized the binding
modes by opening the ligand_out and 6mqf.pdb files.
 3. Results

Figure 1 Analyzed Aspirin using PubChem.

Figure 2 Studied
(Figure 2: Examined the information
the content about
of Aspirin in Aspirin in the Interactions
the Interactions and Pathways
ad Pathways section.
section.)
 Figure 3 Downloaded the 2D structure of aspirin in SDF format from PubChem.

Figure 4 Accessed the PDB page for 6MQF, which was chosen as the protein from the Interactions and Pathways section.
Figure 5 Extended the 2D structure of aspirin, which had been downloaded in SDF format, using Marvin JS.

Figure 6 Demonstration of Aspirin in Marvin Js.

Figure 9 6MQF opened via Autodock Tool.

Figure 11 The grid was configured using AutoDock.

Figure 13 The command prompt was used to switch directories, and the path to the vina.exe file was copied.
Figure 14 The "conf" file was accessed, and code was added next to the copied pathway of vina.exe, followed by initiating the
docking process.

Figure 15 We reviewed the "ligand_out" file generated within our "vina" directory after performing the docking in PyMOL.
 Figure 16 Reviews on Pymol

4. Discussion
The provided data seems to be connected to a study involving molecular modeling or docking,
where different arrangements of a ligand interacting with a target molecule were examined.
Affinity: Across all tested scenarios, the calculated affinity is reported as 0.0 kcal/mol. This
implies that, based on the calculations, there is no favorable binding interaction between the
ligand and the target molecule. An affinity of 0.0 kcal/mol typically indicates that there's no
predicted stable binding or beneficial interaction.
RMSD 1.b.: RMSD (Root Mean Square Deviation) measures how different the predicted
binding position of the ligand (1.b.) is from a reference structure, likely an experimental or
known structure. In all cases, the RMSD values are not zero, indicating that the predicted binding
positions are structurally distinct from the reference structure. Smaller RMSD values suggest
more accurate predictions. For instance, in Mode 6, the RMSD 1.b. is relatively low at 3.555,
implying a better predicted binding position compared to other modes.
RMSD u.b.: RMSD u.b. (Root Mean Square Deviation unbound) is similar to RMSD 1.b. but
measures the structural dissimilarity between the unbound ligand and the reference structure.
Lower RMSD u.b. values also indicate more accurate predictions. For example, in Mode 6, the
RMSD u.b. is 4.054, which is relatively low.
In summary, the data shows that in all the tested modes, there is no predicted favorable affinity
or binding interaction between the ligand and the target molecule. The predicted binding
positions of the ligand, as indicated by RMSD values, differ from the reference structure,
suggesting that the molecular modeling or docking study did not yield a binding arrangement
with a strong affinity for the target molecule in any of the tested modes. This could imply that
the ligand is not well-suited for the target, or that additional factors need to be considered to
enhance the ligand's binding capabilities.
5. Conclusion
To summarize, the study encompassed several stages. It began with an analysis of
Aspirin's characteristics on PubChem, extracting its 2D structure in SDF format, exploring the
RCSB PDB website, and employing the Chemical Sketch Tool. Furthermore, the Aspirin
molecule underwent modifications using Marvin JS, and an examination of proteins associated
with Aspirin was conducted.
During this investigation, a connection between Aspirin and the 6MQF molecule was
identified. However, when the altered Aspirin-6MQF interaction was assessed in PyMOL, no
binding was observed, as indicated by an affinity score of 0. This suggests the necessity for
further research to uncover the reasons behind the lack of binding between the modified Aspirin
molecule and 6MQF. This underscores the need for a more in-depth understanding of their
interaction.

References
 Pickett S.D., Mason J.S., McLay I.M. Diversity profiling and design using 3D
pharmacophores: Pharmacophore-derived queries (PDQ) J. Chem. Inf. Comput. Sci.
1996;36:1214–1232.
 Clark D.E., Pickett S.D. Computational methods for the prediction of "drug-likeness".
Drug Discovery Today. 2000;5:49–58.
 Schreiber S.L. Target-oriented and diversity-oriented organic synthesis in drug
discovery. Science. 2000;287:1964–1969
 Livingstone D.L. The characterization of chemical structures using molecular
properties. A survey. J. Chem. Inf. Comput. Sci. 2000;40:195–209.
 Hertzberg R.P., Pope A.J. High-throughput screening: New technology for the 21st
century. Current Opin. Chem. Biol. 2000;4:445–451.
 Polinsky A. Combichem and chemoinformatics. Curr. Opin. Drug Discovery Dev.
1999;2:197–203
 Dominik A. Computer-Assisted Library Design. In: Bannwarth W., Felder E., editors.
Methods and Principles in Medicinal Chemistry, Vol. 9: Combinatorial Chemistry - A
Practical Approach. Vol. 9. WILEY-VCH; Weinheim, Berlin, New York: 2000. p.
277.