Computer Aided Drug Design
Home>Preclinical Discovery>Computer Aided Drug Design

Computer Aided Drug Design

The molecular modeling platform is an interface between medicinal chemistry and biology with proven record of project specific capability development. Current capabilities are as follows:

Modeling and Designing of Small Compounds

Geometry-optimized 3D models of compounds are generated. 3D models of molecular variants are generated automatically by in-house tool. Novel drug candidates are designed by core replacement, backbone modification etc. of known active compounds.

Generation of geometry-optimized 3D models from 2D structures with the help of Discovery Studio facilities.

Customization of the file format by in-house tools for subsequent use in commercial and in-house tools.

Automatic generation of 3D models of molecular variants around selected molecular cores using fragment libraries by in-house tool AMOLGEN.

Designing of novel small compounds, based on known active compounds as templates, by core replacement, backbone modification etc.

Virtual Screening of Compounds

Virtual screening (structure based and ligand based) of compounds is done in search of active compounds.
More

Molecular docking based virtual screening. Docking tool available:
(a) MOE (commercial)
(b) ASTAC (in-house tool)

Molecular docking based compound profiling against a panel of targets.

Molecular fingerprint based in-house virtual screening approach in search of active compounds.

Customized 3D models of commercially available compounds for virtual screening.

A total of 1.9 million customized 3D models of commercially available compounds obtainable from 17 different vendors.

These compounds are available for virtual screening using Molecular docking.

Evaluation of Binding Potentials of Designed Compounds

The potentials of different designed compounds are assessed by employing different In-Silico approaches.
More

a) Using molecular docking scores.
b) Using Similarity score by structural overlay.

Prediction of ADME-PK Related Properties and Drug-Likeness

Several ADME properties of compounds and Lipinski's Rule of Five (RO5) data are calculated using commercially available tool MOE (Molecular Operating Environment). Also the drug-likeness of a compound is assessed by a number of QSPR based in-house predictive tools and Lipinski's Rule of Five (RO5) using the 3D model of the compound.
More

In-house computational tools based on Quantitative Structure Property Relations (QSPR) have been developed for predicting the following ADME-related physico-chemical properties and drug-likeness of compounds from their 3D structures.
a)LogP
b)LogS
c)LogBB
d)% of Absorption
e)% of Plasma protein binding

Drug-likeness:

a) Overall oral drug-likeness score (DLS).
b) Assessment based on Lipinski's Rule of Five (RO5).

Assessing the hERG Liability of Compounds from Their 3D Structures

The hERG liabilities of compounds are assessed by predicting the pIC50 values of hERG activity by in-house QSAR.
More

The hERG liabilities of compounds are assessed by predicting the pIC50 values of hERG activity based on in-house developed Quantitative Structure Activity Relation (QSAR).

Construction of Homology Model

Homology models of proteins are generated and refined using commercial software MOE.
More

Homology models are used to understand the binding modes of ligands, identifying binding cavity and performing docking based virtual screening in search of novel active compounds.

Recent Publication from the Group

1. Roy S and Sen S. Homology Modeling based solution structure of Hoxc8-DNA complex: role of context bases outside TAAT stretch.
Journal of Biomolecular Structure and Dynamics, 2005, 22, 707-718.


2.
Biswas D, Roy S and Sen S. A Simple Approach for Indexing the Oral Drug-likeness of a Compound: Discriminating Drug-like Compounds from Nondrug-like Ones. Journal of Chemical Information and Modeling, 2006, 46:1394-1401.

3.
Roy S and Sen S. Exploring the potential of complex formation between a mutant DNA and the wild type protein counter part: A MM and MD simulation approach. Journal of Molecular Graphics and Modeling. 2006, 25:158-168.

4. Basu S and Sen S. Turning a Mesophilic Protein into a Thermophilic One: A Computational Approach Based on 3D Structural Features.
Journal of Chemical Information and Modeling. 2009, 49: 1741-1750.


5.
Sinha N and Sen S. Predicting hERG activities of compounds from their 3D structures: Development and Evaluation of a global descriptors based QSAR model. European Journal of Medicinal Chemistry. 2011, 46: 618-630.
Contact Us
Preclinical Discovery:
brlresearch@chembiotek.com

R&D Centers in India:
Kolkata
+91 33 23673151/52
Business Development: India
contact.in@tcgls.com
USA
contact.us@tcgls.com
Europe
contact.eu@tcgls.com
Japan

contact.jp@tcgls.com