Department of Molecular Biotechnology and Health Sciences

In-silico modeling of permeability and ADME properties for drug discovery purposes

My research focuses on the computational prediction of physicochemical properties related to permeability properties and biological activity of molecules.

Permeability and ADME

The permeability of molecules through the biological membranes is a pivotal property in the study of ADME properties of molecules and is investigated following two approaches based on chemometrics and on atomistic level simulations respectively:

Chemometrics approaches (QSPR) are based on the investigation of the relationships between permeability measures (e.g. CACO-2, MTDK, Pampa, etc.) and molecular descriptors using various statistical tools (e.g. Partial Least Squares, PLS, Machine Learning, etc.). A great effort is devoted to developing methods that can furnish information about the mechanisms involved in the passage of compounds through membranes. To do so, we developed the BR Analysis and further investigations are planned to increase the efficacy of the method.

Permeability can also be investigated at an atomic level using Molecular Dynamics simulation. MD simulations generate a time series of configurations of a system that allows the study of a dynamic process such as permeability. MD simulations are carried out in water and apolar explicit solvents (e.g. chloroform) and analyzed using the Markov Model States theory to obtain information about the permeability process. Other simulations based on Molecular Mechanics modeling are evaluated to define the best procedure for different classes of compounds.

The permeability studies are mainly focused on bRo5 classes of compounds, in particular macrocycles, cyclic peptides and, recently, PROTACs.

Biological activity

The biological activity of potential anticancer compounds was studied using ligand- and structure-based virtual screening techniques coupled with MD simulations to reinforce the reliability of the results. Particular attention was devoted to the problem of Protein-Protein Interaction modulation which is a promising field of research in Drug Discovery.

Skills

Molecular Dynamics simulation in different solvents; Quantum mechanics calculations; Molecular Interaction Fields (MIFs) based tools; QSPR based on various statistical approaches (PLS, Machine Learning, BR Analysis, etc.); Ligand- and Structure-based virtual ligand screening.