What is in silico drug design?

In silico drug design is a term that means ‘computer-aided molecular design’, or in other words the rational design or discovery of drugs using a wide variety of computational methods. Especially in the early phases of drug discovery projects, in silico tools based on the knowledge of the target receptor structure (protein-based) or on the chemical structure of active small molecules (ligand-based) are routinely used for discovering and optimizing hit or lead compounds of pharmaceutical interest. Computer-aided molecular design is currently a standard approach that is implemented in all major pharmaceutical and biotechnological research centers. 

What kind of scientific problems are studied at our laboratory?

1. You have a biologically active compound and would like to identify compounds with similar activities?Quantitative structure-activity relationships, or shortly QSAR, are mathematical models describing the relation between a series of chemical structures and their corresponding biological or biophysical responses. Chemical descriptors are at the core of QSAR modeling and many of these have been described, ranging from 1D (e.g. logP or MW), 2D (e.g. molecular topologies), and up to 3D (e.g. the spectrophore technology developed at our laboratory). A classical example of 3D QSAR is known as CoMFA (Comparative Molecular Field Analysis), in which the biological activity of a set of molecules is correlated to their shape and electrostatic properties.

2. You have a crystal structure of a medical important protein and would like to identify molecules that bind to this protein?Molecular docking is one of the most frequently used methods in protein-based drug design, due to its ability to predict the binding-conformation of small molecule ligands to the appropriate target binding site. Two approaches are particularly popular within the molecular docking community. One approach uses a matching technique that describes the protein and the ligand as complementary surfaces. The second approach simulates the actual docking process in which the ligand-protein pairwise interaction energies are calculated.

3. You want to investigate the kinetics of ligand binding?Understanding protein–ligand binding processes is undoubtedly of critical importance in protein structure-based drug design, and much effort is being invested in computational methods to resolve binding. In our lab, we are using large-scale molecular dynamics simulations in combination with Markov state models (MSM) to study kinetic models of binding processes.

4. You want to develop advanced machine learning models to predict the biological activities of novel molecules?Due to the availability of data spanning a much wider chemical space, the use of a large and diverse selection of descriptors, and the development of sophisticated nonlinear machine learning algorithms have increased the generation of global models in recent years. In our laboratory, we are using RDKit and Scikit-Learn to develop and validate sophisticated models to predict biological activities of molecules.

More Information: Prof. Hans De Winter