Model Revision from Temporal Logic Properties in Computational Systems Biology

• Published in: Probabilistic Inductive Logic Programming Vol. 4911; pp. 287 - 304
• Main Authors: Fages, François, Soliman, Sylvain
• Format: Book Chapter
• Language: English
• Published: Germany Springer Berlin / Heidelberg 2008; Springer Berlin Heidelberg
• Series: Lecture Notes in Computer Science
• Subjects: Artificial intelligence; Circadian Cycle; Computer programming / software development; Kinetic Expression; Kripke Structure; Mathematical theory of computation; System Biology Markup Language; Temporal Logic
• ISBN: 9783540786511; 3540786511
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-540-78652-8_11

Systems biologists build models of bio-molecular processes from knowledge acquired both at the gene and protein levels, and at the phenotype level through experiments done in wild-life and mutated organisms. In this chapter, we present qualitative and quantitative logic learning tools, and illustrate how they can be useful to the modeler. We focus on biochemical reaction models written in the Systems Biology Markup Language SBML, and interpreted in the Biochemical Abstract Machine BIOCHAM. We first present a model revision algorithm for inferring reaction rules from biological properties expressed in temporal logic. Then we discuss the representations of kinetic models with ordinary differential equations (ODEs) and with stochastic logic programs (SLPs), and describe a parameter search algorithm for finding parameter values satisfying quantitative temporal properties. These methods are illustrated by a simple model of the cell cycle control, and by an application to the modelling of the conditions of synchronization in period of the cell cycle by the circadian cycle.