puzzle. © shuttlecock

Modelling is everywhere

By Pascale Mollier - Evelyne Lhoste - Catherine Foucaud-Scheunemann, translated by Emma Morton-Saliou
Updated on 01/03/2014
Published on 04/15/2013

Every area of research conducted at INRA – agriculture, nutrition, the environment – uses modelling. Through a wide range of examples, this online report explores how mathematicians, computer scientists and biologists work together to build models that help better understand the complex world around us.

Understanding through modelling

A model is a simplified representation of a complex system or phenomenon, used to measure, understand and predict how that system or phenomenon operates. A set of known or measured input variables are entered into a model, which provides output variables using mathematical, graphic and computer-based processes.
Modelling is a fundamental part of scientific experimentation. Data observed in an experiment is first used to design a model – variables must be linked and transformed into equations – and afterwards, to confirm it.

Modelling in every field of study

Models can be created for all kinds of biological phenomena, from genes to organisms in the case of living creatures, and from plants to landscapes in agri-food systems. Modelling is a component of every area of research at INRA. The examples chosen for this report illustrate the diversity of subjects studied at INRA, including genomes, social sciences and nutrition. Read on to learn about genomic annotation, soil modelling, the spread of pests, the modelling of how salt is released in the mouth, and data mining in social sciences.

Another report will strictly focus on models used to understand and manage agroecosystems, a field of specialty at INRA.

Systems biology, which studies how biological entities such as cells or gut flora function and which relies heavily on modelling, will also be covered in another, separate report.


Mathematics describe the real world in a very special language which can quantify phenomena. Until recently, the use of mathematics in biology was limited to relatively simple calculations, experimental measurements and statistics on a limited amount of data. Today, mathematicians work with biologists to give meaning to enormous swathes of collected data. In other words, they connect the dots between this data and describe how a given system works. Doing so requires minutely breaking down a process into mathematical functions and developing a theoretical model that illustrates what is happening in the system. The validity of the model must then be tested via other experiments.


Modelling is used to illustrate and understand: models are made to describe cell metabolism and the transfer of nitrogen in the environment. It is also used to predict how a system will react to change.

In agriculture, models are used to evaluate the economic, social and environmental effects of crop-growing systems and design new ones, as a support tool in decision-making.

A modelling conference was held during the Forum des Labos at INRA Versailles-Grignon on 16 November 2012. Click below to watch the keynote address (in French) by Michaël Chelle, a member of INRA AgroParisTech’s Environment and Arable Crops Joint Research Unit:

Scientific contact(s):

Associated Unit(s):
UMR1091 EGC Environment and Arable Crops Joint Research Unit

Long live algorithms!

“Natural algorithms are the language of the living world. The only one capable of rendering its descriptive complexity”.

Click here to watch the conference given by Bernard Chazelle on 18 October 2012 at the Collège de France in Paris.

Bernard Chazelle, Eugene Higgins Professor of Computer Science at Princeton University, held the Chair in Information Technology and Digital Sciences at the Collège de France in 2012.

A collective work

This book, entitled "Analyse de sensibilité et exploration de modèles - Applications aux sciences de la nature et de l'environnement" helps to choose the most adapted method for analysing models in various disciplines: oceanography, environment, water management, ecology, agronomy, etc. It is aimed at scientists and users of models.

The authors are almost all members of the interinstitutional research network MEXICO (Methods for EXploring the Informatics of COmplex models).

Several authors are from INRA. Coordination: Robert Faivre, Bertrand Iooss, Stéphanie Mahévas, David Makowski, Hervé Monod. In French.

Editions Quae, collection Savoir-faire, 2013, 352 pages, price: 55 euros. Available in bookshops in France and on the Quae website.