Strumenti in silico per la predizione del passaggio della barriera ematoencefalica.

The blood-brain barrier (BBB) prevents solutes in the circulating blood from crossing into the extracellular fluid of the central nervous system. However, there are two main ways of crossing the BBB, namely passive transmembrane diffusion, and carrier-mediated transport. Some molecules, especially lipophilic ones, are able to cross the BBB by passive diffusion. Some others may be substrates or inhibitors of transmembrane transporters, with the efflux transporter P-glycoprotein (P-gp) being the most studied. The BBB represents a major obstacle to the delivery of drugs to the CNS. It is estimated that at least 98% of all drug candidates developed for treatment of CNS diseases never reach the clinic. This high attrition rate can be reduced at early development phases through in silico tools able to predict ADME properties of the compounds, and thus BBB permeability. This thesis’ main goal is to test several open access online webservers that predict BBB permeability, by considering tools focusing on the two main mechanisms of BBB crossing, i.e. passive diffusion and P-gp efflux. Using appropriate datasets of compounds retrieved from the literature, the features of each tool have been tested and the accuracy of the predictions have been statistically evaluated to identify the most reliable tools. Lastly, the WEKA software was used to generate classification models based on the physicochemical properties to model BBB permeability.