MetaPred: Prediction of Cytochrome P450 Isoform Responsible for Metabolizing a Drug Molecule

Overview

MetaPred is a computational platform developed for predicting the cytochrome P450 (CYP450) isoform responsible for metabolizing a drug molecule.

The study focused on five major CYP isoforms:

• CYP3A4
• CYP2D6
• CYP1A2
• CYP2C9
• CYP2C19

These isoforms metabolize most therapeutic drugs in humans.

---

Research Article

Title: Prediction of cytochrome P450 isoform responsible for metabolizing a drug molecule

Authors: Nitish K Mishra, Sandhya Agarwal, and Gajendra P. S. Raghava

Journal: BMC Pharmacology (2010)

DOI: https://doi.org/10.1186/1471-2210-10-8

https://doi.org/10.5281/zenodo.20084455

---

Background

Cytochrome P450 enzymes are involved in drug metabolism and biotransformation.

Predicting the CYP isoform responsible for metabolizing a drug is important for:

• ADMET analysis
• Drug discovery
• Toxicity studies
• Drug-drug interaction analysis
• Pharmacokinetic profiling

---

Dataset

Main Dataset

• Total molecules: 216
• CYP3A4 substrates: 111
• CYP2D6 substrates: 47
• CYP1A2 substrates: 29
• CYP2C9 substrates: 20
• CYP2C19 substrates: 19

Independent Dataset

• Total molecules: 146

---

Descriptor Calculation

The study used:

• TSAR
• ADMEWORKS
• Chemistry Development Kit (CDK)

A total of 2392 molecular descriptors were initially calculated.

After feature selection, 41 descriptors were used for SVM model development.

---

Machine Learning Methods

The following algorithms were tested:

• Support Vector Machine (SVM)
• Random Forest
• BayesNet
• Logistic Regression
• Multilayer Perceptron

SVM models achieved the best performance.

---

Model Performance

Isoform	Accuracy (%)	MCC
CYP3A4	81.42	0.63
CYP2D6	81.24	0.54
CYP1A2	83.19	0.49
CYP2C9	84.51	0.40
CYP2C19	70.80	0.15

Overall Performance

• Average Accuracy: 86.02%
• Overall Accuracy: 82.81%

---

Web Server

MetaPred Web Server:

http://crdd.osdd.net/raghava/metapred/

Supported inputs:

• MOL2
• SDF
• SMILES
• JME molecular editor

---

Applications

MetaPred can be used for:

• Drug metabolism prediction
• Virtual screening
• ADMET analysis
• Drug discovery
• Medicinal chemistry research

---

Technologies Used

• Linux
• WEKA
• SVM_light
• CDK descriptors
• CGI-Perl

---

Citation

Mishra NK, Agarwal S, Raghava GPS. Prediction of cytochrome P450 isoform responsible for metabolizing a drug molecule. BMC Pharmacology. 2010;10:8.

---

Contact

Prof. Gajendra P. S. Raghava
Institute of Microbial Technology
Chandigarh, India

Email: raghava@iiitd.ac.in

---

Generated from the uploaded MetaPred paper.