One of the research lines of the Costanzi Research group focuses on the application of computational chemistry to the study of the interaction of chemicals with living organisms. We are particularly interested in: a) the development, optimization and/or testing of computational research strategies; and b) the application of computational methodologies to solve specific, contingent problems, for instance the identification of chemicals that can modulate the activity of a given target of interest.

Among biological targets, we are particularly interested in G protein-coupled receptors (GPCRs).

Chemicals and living organism. Biologically active molecules are naturally occurring or synthetic chemicals that interact with leaving organisms affecting them in either a positive (e.g. pharmaceutivals) or negative manner(e.g. chemical warfare agents or toxins).

The great majority of biologically active molecules exert their actions by binding to specific structures located in their target cells – quoting German immunologist and Nobel Prize winner Paul Ehrlich, “corpora non agunt nisi ligata”, or bodies are not active unless they bind to something.

Hence, to understand how biologically molecules chemicals regulate and alter physiological functions, discover new biologically active molecules, or disrupt the effect of those that are harmful, it is of fundamental importance to study the way these compounds interact with their biological targets.

Computational Chemistry. With these premises in mind, we conduct computational modeling research to produce models that can explain and forecast:

• the structure and the functioning of the cellular targets of biologically active molecules;
• the nature and the strength of the interactions between the chemicals and their targets;
• the molecular properties of biologically active molecules in relation to their activity profile.

To this extent, among several others, some of the computational techniques that we use in our research include:

• Homology modeling, to build three-dimensional models of targets for which experimental structures are not available;
• Molecular docking and virtual screening, to afford three-dimensional models of biologically active chemicals bound to their targets or screen thousands or millions of compounds to infer their likelihood of binding to a given target;
• Quantitative Structural Activity-Relationship (QSAR) techniques, to correlate the physicochemical and topological properties of biologically active chemicals with their activity and yield predictive models for the design of new molecules.