Historically, mathematics has been used extensively in the sciences to describe, explain, and ultimately predict the behaviour of complex systems. Starting from the seventies, models have been used widely also to study complex biological systems and to understand the fundamental biological mechanisms. Mathematical, computational, theoretical modelling approaches have been used for example to represent hypotheses (i.e., circulation in physiology), to test theories, to make predictions (i.e., pharmacological response to a drug), to design outcomes (i.e., intake determination for optimal growth), and to analyse data. Independently from the specific application and from the specific modelling/computational techniques, the common denominator of this area is the use (and the integration) of methodologies/tools coming from the statistics, the mathematics and the artificial intelligence to derive qualitative information about the main variables of the biological system or to make quantitative prediction of the phenomena under investigation. The research in this field requires the integration of different competences ranging from the dynamic system theory to the stochastic processes, from the numerical analysis to the statistics, from the medicine to the biology.

In this context, our research activities are mainly focused on the ordinary differential equations and on the compartmental models, on the nonparametric modelling of temporal profiles and surfaces, and, more recently, on the gene regulatory networks.

A particular attention is devoted to the use of stochastic models and to the Bayesian approaches and Markov Chain Monte Carlo methods for the quantitative estimation of the parameter of a model into a framework characterised by different sources of uncertainty. Classical problems related to the identification of parametric models, as the a-posteriori identifiability, are tackled in different contexts. Some of the developed applications have required the definition of pharmacokinetic and pharmacodynamic models. Other ones have required the development of in vitro/in vivo models for an early evaluation of the potency of an antitumor agent during the development process of the drug. Moreover, the analysis of data coming from a population of subjects and the consequent modelization of the inter-individual variability is one of the most interesting topics of our research. The estimation of non measurable biological signal by using deconvolution stochastic techniques is a further topic on which we are working.

Please find further information searching in our publication repository or by contacting the Responsible of each area.

People working on the topic:
    Nadia Terranova, Paolo Magni

External Collaborations:
   Prof. G. De Nicolao - Dipartimento di Informatica e Sistemistica, Universita' degli Studi di Pavia, Italy.
   Prof. C. Cobelli, Prof. G. Toffolo, Prof. G. Sparacino - Dipartimento di Ingegneria dell'Informazione,
    Universita' degli Studi di Padova, Italy.
   M. Rocchetti, M. Germani - Nerviano Medical Science (MI), Italy.
   I. Poggesi - Glaxo, Verona, Italy.