Raúl Guantes BioDynamics Lab
Biological networks control an organism survival and function both at the cellular level (genetic networks) and at a global level(neural networks). As such, they do not work in isolation but are highly interconnected and embedded in a continuously fluctuating and changing environment. Despite the wealth of signals and external variations they have to deal with, they seem to have been exquisitely tuned by evolution to reliably perform specific functions. This leads to some broad questions that we try to address in the lab at different levels:
How can biological networks integrate and process information from different signals? How can they operate robustly in the presence of noise and undesirable fluctuations? What are the mechanisms underlying adaptation to environmental context? What is the relation between structure and function of simple biological networks?
The advances in experimental techniques and the availability of vasts amounts of data is complementing the traditional molecular approach to research in Biology with a Systems perspective, in which one tries to understand the functioning of a whole decomposing it into simpler 'modules', similar to the engineer design of complex devices. Questions can here be answered at a more quantitative level, and the combination of experiment and mathematical modeling is being proven very fruitful. While this approach has been followed by some time in Neuroscience, it is only recently becoming widely applied in Molecular and Cell Biology. It turns out that one can trace interesting parallelisms in the ways neural and genetic networks process information.
We use a combination of mathematical models, numerical simulation and theoretical techniques to investigate some of the above issues in relevant biological circuits. Since networks are not static entities, one of our main tools is non-linear dynamics, together with the theory of stochastic processes. Statistical physics, signal detection or information theory are also applied. We also establish collaborations with experimental groups where theoretical models can be supported by quantitative data.