Computational Sciences

Life sciences, just as with material or environmental sciences, address increasingly complex processes that could be mastered through the computational integration of multi-physical and multi-scale models.

Thus, the study and design of new hardware requires phenomena to be modeled and simulated, from the nanoscopic to large-scale structures, including their interactions and overlays. INRIA will focus on modeling and multi-scale computational techniques of hardware in nano-simulation. To be able to accomplish this, data originating very different measuring devices must be able to be integrated, and the reliability and relevance of simulations must be able to be evaluated. In addition, a dedicated solution algorithm will need to be developed on large-scale interactive calculation grids.

Multi-physical and multi-scale integration is also necessary in biology, where one of the Institute’s objectives is to contribute to the development of a computational cell that reports the exchange of energy and signals between cells, as well as its internal operation. For example, this would involve gene expression in proteins and the gene interaction networks studied in genomics. To do so, it is most particularly necessary to find new algorithm and computer methods to integrate the data provided by new technologies (DNA chips, chromatin, etc.), to describe the rules of elementary interactions, and to model and simulate their dynamics.

« Computational models of the environment help to analyze scenarios and evaluate the risks in order to deploy strategies of prevention and adaptation. »

Another objective is the agronomic and biological model of the computational plant that combines agronomics with biology, applied mathematics, automation, graphic computing, geometry and combinatorics. The study of phytoplanctons and micro-algae will require developing realistic models of interactions between species and substrates, and designing control strategies for effective depollution and optimization of energy production.

In computational ecology, a particularly significant inter-disciplinary field, the Institute desires to integrate models of animal populations at various trophic levels in order to study how an ecological system responds to natural or human aggression, and to be able to plan effective conservation measures. Microbial ecological models are very important for numerous applications, including treating and de-polluting water and soils. The Institute would like to contribute to studying the biosphere, but the associated issues are significant: greenhouse-effect gas, global warming, pluviometry, freshwater, desertization, etc. Computational models of the environment help to analyze scenarios and evaluate the risks in order to deploy strategies of prevention and adaptation. There are numerous obstacles to completing these models, including problems relating to observation, data assimilation, prediction and follow-up, very variable time scales, and numerous interdependent processes. These are all fields in which INRIA has solid expertise.