Rational Drug Design using Genetic Algorithm

Final Year project
Rational Drug Design Using Genetic
Algorithm
Case of Malaria Disease

Supervision by
Assoc.Prof.Imad Fakhri Taha Alshaikhli
Presented By
Hassen Mohammed Abdullah
Alsafi
International Islamic University Malaysia

Agenda

• Introduction.

• Problem statement.

• Objectives.

• Proposed methods.

• Findings and Analysis.

• Challenges and Difficulties faced.

• Conclusion and Future work.
2
Hassen Alsafi International Islamic University Malaysia 1

Introduction

 How a drug works and how we can expect the body
to respond to the administration of a drug?
 Drug design is known as approach uses specifics
tools to explore and search for the best drug
candidate.

Drug Compound
Hassen Alsafi
Protein Medicine 3 3
International Islamic university Malaysia

problem statement

 What is the best drug candidate for x disease ?
 Drug design and discovery  take years for
discovering a new drug and very costly.
 Effort  to cut down the research timeline and cost
by reducing laboratory experiment  use
computational computer modeling.

4

Rational drug design approach(rdda)

Foundation of drug design and discovery.
Answer the question , which molecule fit best
to the protein active site?

Computational Molecular Docking (CMD)

5

Objectives

1. Find and Select the target disease in the human
body.(e.g malaria)

2. Search and choose the best drug candidate.

3. Conduct computational drug design simulation.

4. Propose some drugs against certain disease
based on results.

6

Drug design and development process

7

Genetic algorithm flowchart

8

Proposed methods

1. Target selection and identification.
1.1 Protein preparation in ADT
1. Drug or ligand identification.
2.1 Ligand preparation in ADT
1. Perform the molecular docking simulation.
2. Techniques used in docking algorithm.
3. Evaluation .
9

Methodology
Computational Molecular docking

Ligand database Target Protein

AutoDock 4.2 Molecular docking

Ligand docked into protein’s active site

10

AutoDock 4.2

 Automated computational molecular docking
programs .

 It is designed to predict how small molecules, bind
to a receptor of known 3D structure.

 It uses Genetic Algorithm (GA) .

11

AutoDock 4.2

12

Methods and materials
1. Target selection and identification .
Target disease Target protein

Malaria 2GHU.pdb

 The protein 3D structured was retrieved form
RCSB database.

13

Autodock workflow

14

Autodock proposed Framework

15

Protein databank (pdb)
 Molecular protein repository .
 Contains a tons of protein stored in the repository.
 In order to convert the drug compound from .sdf to
pdb <openbabel> software used by the following
commend line:

 -i: input type(i.e .sdf and pdb)
 -o: output(convert) type
16

Grid file parameters(gfp)

 After finish the preparation of protein and

drug , now the task is to precalculate the grids
using the following Linux commend line:
autogrid4 –p filename.gpf –l
filename.glg

-p: used to specifics the grid parameter file
gpf: grid parameters file
–i: used as log file output 17

Grid file parameters(gfp)

18

Docking file parameters in adt

 Primary goal of AutoDock is to instruct the drug to
move inside the space search grid.
 GA selected as search algorithm in the experiment.
 Run the following Linux commend line :

autodock4 –p filename.dpf –l filename.dlg

19

Experiment results
 Setup the environment

20

Equipments used in the experiment

21

Tools and materials

22

Genetic algorithm in autodock
 ADT represent chromosome as a vector of real
number .

Quaternion genes
Tx Ty Tz Qx Qy Qz Qw R1 Rn

Translation genes
GA features in ADT:
1. Solution space.
2. Genetic code (chromosome)
3. Genetic operations
4. Fitness function
23

Results and discussion
 Experiment conduct of 3 cases.
 Case 1 : Default parameters.
 Case 2 : Parametric study.
 Case 3: Computational Docking Time (CDT).

24

Case 1 : default parameters
 Run CMD in 20 drugs compound with 1
target protein.

25

[1]
Log p: octanol/water partition coefficient


26


27


28

[1]
Log p: octanol/water partition coefficient


29

Case 2 : Parametric study
 480 samples has been investigated with
different parametric value.
Parameter Value

Pop size(50) 50,100,150

Crossover rate(0.2) 0.2, 0.4, 0.6, and 0.8

Mutation(0.01) 0.01 and 0.02

30


31


32

se 3 : computational docking time

33

Challenges faced
 Compiling the python source code under ADT
environment.
 Installing the openbabel software.
 Dealing with the bioinformatics tools.
 Time given to complete the project.
 Moving from the old building to the new building 

34

Conclusion and future work
 Computational molecular docking with GA are
crucial tools in RDD.
 Using the ADT we can reduce the use of
laboratory experiments(but not at all)
 RDD helps to reduce the time required to design
and discover new drugs .
 Future work
 Further investigation is needed to select the best
potential drug candidate .
 I propose to deploy the grid computing in the
CMD. 35

Conclusion and future work

 In order to perform the CMD faster and accurate ,
the high speed computers is needed.

36

Acknowledgments
Special thanks to My beloved supervisor

Assco.Prof.Dr.Imad Fakhri Taha Alshaikhli
37

Thank you for your attention
Q&A

38

Rational Drug Design using Genetic Algorithm

More Related Content

Viewers also liked (10)

Recently uploaded (20)

Rational Drug Design using Genetic Algorithm

Editor's Notes