In silico prediction of drug targets
Understanding biological processes in health and disease requires elaboration of complex signalling networks. It is, therefore, desirable before starting to experimentally unravel the complex interplay among various molecules to be able to model and predict such interactions in the system of choice. Seeking to address this issue, the EU-funded project ‘Validated predictive dynamic models of complex intracellular pathways related to cell death and survival’ (VALAPODYN) developed a systems biology approach to model the dynamics of molecular interaction networks (MINs) related to neuronal death and survival. In particular, the consortium was interested in performing in silico simulations to identify new potential drug targets to treat neurodegeneration. To set up the integrated database, project partners collected all relevant biological data and built a network which reflected the biological processes which were to be modelled. Known signalling cascades and novel biological data were subsequently used to train the dynamic model at selecting therapeutic targets from different simulation conditions. The output of this model was a selection of putative therapeutic targets which were subsequently validated at the biological level in the system of choice. As a proof of principle, project partners tested the MIN model on induced neurodegeneration in a mouse model and performed RNA and protein analysis in various segments of the brain. The experimental data was subsequently fed into the MIN model which described the behaviour of a large signalling network including 521 molecules (genes/proteins) connected by 3,069 direct and oriented interactions.Overall the MIN model was estimated to be one of the largest dynamic signalling models with a relative error of 16 %. Neurodegeneration simulation identified three potential therapeutic targets (BDNF, Vala09, Vala01) based on the ability to inhibit their action and potential to act as a therapy. The VALAPODYN innovative dynamic modelling of molecular networks constitutes an invaluable tool for researchers and holds great promise for the elaboration of various disease states.
