Computational Resources for QSAR

Ligand-similarity based lead identification is a technique which follows the principle of similarity. It does not require information about 3D structure of the target protein. It is assumed that molecules having similar structure will have similar chemical properties. Hence, the information provided by a compound, or set of compounds known to bind to the desired target is used to identify new compounds from the external databases of chemical compounds using virtual screening approaches. The most commonly used methodology for ligand-similarity based lead identification is as follows:

Quantitative Structural Activity Relationship (QSAR) process quantitatively correlates structural molecular properties (descriptors) with functions (i.e. physicochemical properties, biological activities, toxicity, etc) for a set of similar compounds. It uses linear statistical methods such as Multiple Linear Regression, Partial Least Square, or non-linear methods like Support Vector Machines (SVM), Artificial Neural Network (ANN), Decision Trees, Bayesian Classifier, etc., to generate a mathematical model that connects experimental measures with a set of chemical descriptors. The main objective of QSAR models is to allow the prediction of biological activities of untested or novel compounds to provide insight into relevant and consistent chemical properties or descriptors (2D/3D) which defines the biological activity. Once, a series of predicted models are collected, these can be used for database mining for the identification of novel chemical compounds, particularly, for those having drug-like properties (following Lipinski’s Rule of Five) along with suitable pharmacokinetic properties.

Web Servers/Databases/Mirror Sites:

Web Interface on Libraries:

Standalone Software:

Links: