|The PharmacoInformatics group is devoted to the development and application of computational methodologies in the area of drug design and development.
Nowadays, computational methodologies are widely applied in many steps of the drug discovery and development; from the structural modeling of a pharmacological target to the prediction of the ligand binding affinity. However, in the vast majority of cases the limitations of the current technology allow only to obtain approximate representations of the complex biological phenomena that are the subject of interests in the development of new drugs.
Our group aims to improve the current state-of-the-art with a pragmatic approach. We want to develop useful tools that increase the efficiency of the pharmaceutical R&D process. At the same time, the need of producing robust models led us to overcome reductionist approaches and to develop multi-scale methods, depicting richer and more realistic representations of the phenomena under study than those produced by classical computational methods.
Main Research Lines
In silico methods for drug safety assessment
Drug safety assessment is one of the most important bottlenecks in today’s drug development. Computational (in silico) methods offer a very interesting alternative to other experimental methods for ealy drug safety screening; they are faster, cheaper and require no amount of your valuable compound to obtain a prediction that sometimes has comparable quality with those obtained using high-throughput in vitro methods.
In this area we are coordinating the project eTOX, a public-private partnership within the framework of the European Innovative Medicines Initiative Joint Undertaking (IMI-JU) aiming to produce in silico predictions of in vivo toxicity endpoints for novel drug candidates.
Read more… eTOX
Molecular Interaction Fields in drug design
Molecular Interaction Fields (MIF) have many applications in drug design; from the characterization of the receptor binding site, to the comparative analysis of ligand series, including their use as highly relevant molecular descriptors.
MIF are the core of most of the methodologies used in our group, either to characterize ligands, target binding sites or to obtain high quality molecular descriptors. In particular, we used the MIF generated using the program GRID (Molecular Discovery Ltd.) due to the accuracy of the results, its ease of use and the speed of computation.
Novel MIF-based molecular descriptors
An important line of research is devoted to the development of MIF-based molecular descriptors (MD). In particular, we are developing a new generation of GRid INdependent Descriptors (GRIND) and implementing them in the software Pentacle.
These MD are highly suitable for 3D-QSAR modelling and virtual screening and, due to their property of being alignment invariant, they can be computed and applied in minutes, without the need of carrying out time-consuming structural alignment.
Read more… Pentacle
Molecular modeling of GPCR
G-protein coupled receptors (GPCR) are the target of most drugs in the market. However, since they are membrane proteins, only a handful of X-ray crystallography structure has been reported and the homology modeling on these proteins is still challenging.
The strategy used in our group to work with these receptors is to combine direct (homology modeling, SBDD) and indirect (3D QSAR) approaches in a synergistic way.
Read more… Biopsycho