Inference of ecological networks and possibilistic dynamics based on Boolean networks from observations and prior knowledge
2025
Paulevé, Loïc | Gaucherel, Cédric | Laboratoire Bordelais de Recherche en Informatique (LaBRI) ; Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS) | Botanique et Modélisation de l'Architecture des Plantes et des Végétations (UMR AMAP) ; Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche pour le Développement (IRD [Occitanie])-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Université de Montpellier (UM) | ANR-20-CE45-0001,BNeDiction,Ensembles de Réseaux Booléens Prédictifs(2020) | ANR-23-CE45-0008,REBON,Abstraction des Réseaux de Réactions vers des Réseaux Booléens pour Améliorer l'Inférence et le Contrôle en Biologie des Systèmes(2023)
International audience
Показать больше [+] Меньше [-]Английский. Being able to infer the interactions between a set of species from observations of the system is of paramount importance to obtain explanatory and predictive models in ecology. We tackled this challenge by employing qualitative modelling frameworks and logic methods for the synthesis of mathematical models that can integrate both observations and expert knowledge on the system. Boolean networks is a qualitative modelling framework, which enables reasoning exhaustively on possible dynamics of the system. After devising a formal link between ecological networks and the causal structure of Boolean networks, we applied a generic model synthesis engine to infer Boolean models that are able to reproduce the observed dynamics of a protist community and of a planktonic ecosystem. Our inference method supports optimization criteria to derive the most parsimonious and most precise models. It is also able to integrate prior knowledge on the ecological network, adding constraints on impossible interactions, which is necessary to obtain realistic predictions. Such constraints may, however, prove to be too strict, in which case our method is able to conclude on the absence of a model compatible with both the observations and the input hypotheses. We demonstrated our methodology on experimental data of a protist community and of a planktonic ecosystem and showed in each case its ability to recover essential and sufficient ecological interactions to explain the observed dynamics.
Показать больше [+] Меньше [-]Ключевые слова АГРОВОК
Библиографическая информация
Эту запись предоставил Institut national de la recherche agronomique