Structural bioinformatics represents a section of bioinformatics dealing with analysis and prediction of three-dimensional (3D) structures of biological macromolecules such as proteins, RNA, and DNA. Through skilled simulative techniques, involving the use of 3D structures, it is possible to compare overall folds or local motifs, to study the principles of folding and evolution, to analyze binding interactions, molecular recognition and structure/function relationships of large macromolecules. Structural bioinformatics requires the use of experimentally determined structures or specially made computational models, and can be seen as a part of computational structural biology.
In our group various structural bioinformatics tools are used to plan experiments, to organize and verify experimental data and to control the systematic design of protein mutants with altered functional properties.
In details our research area concerns:
- Investigation of the dynamical features of proteins and nucleic acids through classical molecular dynamics simulation or enhanced sampling simulation techniques. These simulative techniques are used to compare time evolutions of structural properties of wild-type and mutated enzymes, to correlate their structure-dynamics-functions relationship in order to obtain a complete picture of the acquired altered functionality from an atomistic point of view.
- Structural characterization, design and construction of large DNA assemblies molecular models, experimentally obtained through covalently linked DNA oligonucleotides or origami techniques, to be used for cargo delivery of active molecules for biomedical purposes. We use large computational facilities to carry out hundred of nanoseconds simulations of this nano-biomaterial systems that include millions of atoms.
- Investigation of electrostatic interactions in proteins through simulative methods. Through the calculation of the electrostatic potential distribution around macromolecules we study protein-ligand or protein-protein interaction mechanisms.
- Investigation of proteins-ligand and protein-protein molecular recognition through molecular docking simulations. Various programs and algorithms are used to detect information on the way ligands interact with receptors, for biochemical or pharmacological purposes.
- Structure-based virtual screening, using computer-based methods and extended virtual compounds databases, to discover new ligands on the basis of biological structures. This procedure can be performed at low resolution, using millions of compounds, and then translated into a better definition of the obtained results through higher resolution methodologies.
- Creation of novel inhibitory compounds obtained on the basis of previous biochemical data.
- Prediction of protein tertiary structures through homology-based approaches, protein threading and refined combined approaches. The structural models can provide valuable indications in the absence of structures obtained through experimental methods.
We are open to any form of collaboration with non-profit institutions or private companies.