Equation discovery for systems biology: finding the structure and dynamics of biological networks from time course data

Current Opinion in Biotechnology

Volumn 19, Issue 4, 2008, Pages 360-368

  Džeroski, Sašo ^a   Todorovski, Ljupčo ^b  

^a JO EF STEFAN INSTITUTE   (Slovenia)

^b UNIVERSITY OF LJUBLJANA   (Slovenia)

Author keywords

[No Author keywords available]

Indexed keywords

ARTIFICIAL INTELLIGENCE; DYNAMICS; LEARNING SYSTEMS;

BIOLOGICAL NETWORKS; DOMAIN KNOWLEDGE; DYNAMIC BEHAVIORS; EQUATION DISCOVERY; MACHINE-LEARNING; NETWORK STRUCTURES; PARTIAL KNOWLEDGE; SCIENTIFIC DISCOVERIES; SIGNALING NETWORKS; STRUCTURE AND DYNAMICS; SYSTEMS BIOLOGY; UNKNOWN NETWORKS;

KNOWLEDGE BASED SYSTEMS;

BIOLOGICAL MODEL; KNOWLEDGE; LEARNING; MACHINE LEARNING; MATHEMATICAL MODEL; MOLECULAR BIOLOGY; MOLECULAR DYNAMICS; PRIORITY JOURNAL; PROCESS MODEL; REVIEW; SIGNAL TRANSDUCTION; SYSTEMS BIOLOGY;

MODELS, THEORETICAL; SYSTEMS BIOLOGY;

EID: 49549123570     PISSN: 09581669     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.copbio.2008.07.002     Document Type: Review

References (40)

1. Price N.D., and Shmulevich I. Biochemical and statistical network models for systems biology. Curr Opin Biotechnol 18 (2007) 365-370
2. Langley P. Elements of Machine Learning (1996), Morgan Kaufmann, San Francisco
3. Mitchell T.M. Machine Learning (1997), McGraw Hill, New York
4. Langley P., Simon H.A., Bradshaw G.L., and Żytkow J.M. Scientific Discovery (1987), MIT Press, Cambridge, MA
5. Džeroski S, Todorovski L. (Eds.): Computational Discovery of Scientific Knowledge. Berlin: Springer; 2007.An overview of the state-of-the-art in computational scientific discovery, including several approaches to equation discovery and applications to finding the structure and dynamics of biological networks.
6. Džeroski S., and Todorovski L. Discovering dynamics: from inductive logic programming to machine discovery. J Intell Inf Syst 4 (1995) 89-108
7. Russel S., and Norvig P. Artificial Intelligence: A Modern Approach. edn 2 (2003), Prentice Hall, Upper Saddle River, NJ
8. Pearl J. Heuristics: Intelligent Search Strategies for Computer Problem Solving (1983), Addison-Wesley, New York
9. Todorovski L, Džeroski S, Džeroski S, Todorovski L: Computational Discovery of Scientific Knowledge. Edited by Džeroski S, Todorovski L. Berlin: Springer; 2007:69-97. Describes how grammar-based equation discovery can be used to take into account different types of domain knowledge, including process-based domain knowledge about basic processes that govern the dynamics of systems in the area of study.
10. Todorovski L., Ljubič P., and Džeroski S. Inducing polynomial equations for regression. Lect Notes Comput Sci 3201 (2004) 441-452
11. Džeroski S., Todorovski L., and Ljubič P. Using constraints in discovering dynamics. Lect Notes Comput Sci 2843 (2003) 297-305
12. Pečkov A., Džeroski S., and Todorovski L. A minimal description length scheme for polynomial regression. Lect Notes Comput Sci 5012 (2008) 284-295
13. Grünwald P.D. The Minimum Description Length Principle (2007), MIT Press, Cambridge, MA
14. Pečkov A, Džeroski S, Todorovski L. Multitarget polynomial regression with constraints. In Proceedings of the ECML/PKDD International Workshop on Constraint-Based Mining and Learning, Edited by Nijssen S, de Raedt L. Warsaw, Poland: Warsaw University; 2007:61-72.
15. Tyson J.J., Chen K., and Novak B. Network dynamics and cell physiology. Nat Rev Mol Cell Biol 2 (2001) 908-916
16. Goryachev A.B., and Pokhilko A.V. Computational model explains high activity and rapid cycling of Rho GTPases within protein complexes. PLOS Comp Biol 2 (2006) 1511-1521
17. Todorovski L, Džeroski S. Declarative bias in equation discovery. In Proceedings of the Fourteenth International Conference on Machine Learning, Edited by Fisher DH. San Mateo, CA: Morgan Kaufmann; 1997:376-384.
18. Gennemark P., and Wedelin D. Efficient algorithms for ordinary differential equation model identification of biological systems. IET Syst Biol 1 (2007) 120-129. An approach to equation discovery designed with finding the structure and dynamics of biological networks in mind; similar to, but less formalized than grammar-based equation discovery.
19. Arkin R.P., and Ross J. Statistical construction of chemical reaction mechanisms from measured time-series. J Phys Chem 99 (1995) 970-979
20. Bunch D.S., Gay D.M., and Welsch R.E. Algorithm 717: subroutines for maximum likelihood and quasi-likelihood estimation of parameters in nonlinear regression models. ACM Trans Math Soft 19 (1993) 109-130
21. Dennis J.E., Gay D.M., and Welsch R.E. Algorithm 573: NL2SOL-an adaptive nonlinear least-squares algorithm. ACM Trans Math Soft 7 (1981) 369-383
22. Atanasova N., Todorovski L., Džeroski S., Rekar Remec S., Recknagel F., and Kompare B. Automated modelling of a food web in lake Bled using measured data and a library of domain knowledge. Ecol Model 194 (2006) 37-48
23. Džeroski S, Todorovski L, Magnani L, Nersessian NJ, Pizzi C: Encoding and using domain knowledge on population dynamics for equation discovery. Logical and Computational Aspects of Model-Based Reasoning. Edited by Magnani L, Nersessian NJ, Pizzi C. Dordrecht: Kluwer; 2002:227-247.
24. Todorovski L., and Džeroski S. Using domain specific knowledge for automated modeling. Lect Notes Comput Sci 2810 (2003) 48-59
25. Todorovski L., and Džeroski S. Integrating knowledge-driven and data-driven approaches to modeling. Ecol Model 194 (2006) 3-13. Inducing process-based models through grammar-based equation discovery, demonstrating the utility of domain knowledge in environmental applications.
26. Voit E.O. Computational Analysis of Biochemical Systems (2000), Cambridge University Press, Cambridge, UK
27. Langley P, Sanchez J, Todorovski L, Džeroski S. Inducing process models from continuous data. In Proceedings of the Nineteenth International Conference on Machine Learning, Edited by Sammut C, Hofmann A. San Mateo, CA: Morgan Kaufmann; 1997:347-354.
28. Bridewell W., Langley P., Todorovski L., and Džeroski S. Inductive process modeling. Mach Learn 71 (2008) 132. Introduces a formalism for representing process-based models and domain knowledge, and a learning method that searches the space of process-based models directly.
29. Langley P., Shiran O., Shrager J., Todorovski L., and Pohorille A. Constructing explanatory process models from biological data and knowledge. Artif Intell Med 37 (2006) 191-201. A successful application of the process-based modeling approach to the task of constructing a biological network model from time course data, describing the dynamics of glycolysis.
30. Torralba A., Yu K., Shen P., Oefner P., and Ross J. Experimental test of a method for determining causal connectivities of species in reactions. Proc Natl Acad Sci 100 (2003) 1494-1498
31. Todorovski L, Bridewell W, Shiran O, Langley P. Inducing hierarchical process models in dynamic domains. In Proceedings of the Twentieth National Conference on Artificial Intelligence, Edited by Veloso MM, Kambhampati S. Pittsburgh, PA: AAAI Press; 2005:892-897.
32. Garrett SM, Coghill GM, Srinivasan A, King RD: Learning qualitative models of physical and biological systems. Computational Discovery of Scientific Knowledge. Edited by Džeroski S, Todorovski L. Berlin: Springer; 2007:248-272. Describes an approach to learning qualitative models of dynamic systems and illustrates its relevance to finding biological networks on the glycolysis pathway.
33. Zupan B, Bratko I, Dems? ar J, Juvan P, Kuspa A, Halter JA, Shaulsky G: Discovery of genetic networks through abduction and qualitative simulation. Computational Discovery of Scientific Knowledge. Edited by Džeroski S, Todorovski L. Berlin: Springer; 2007:228-247.
34. Zapatka M., Koch Y., and Brors B. Ontological analysis and pathway modelling in drug discovery. J Pharm Med 22 (2008) 99-105. Addresses the task of identifying significantly altered pathways in genome-wide data, discusses pathway representations for mathematical modelling and pathway analysis methods, and gives some examples of the successful modelling of signal transduction pathways.
35. Ideker T., Thorsson V., Ranish J.A., Christmas R., Buhler J., Eng J.K., Bumgarner R., Goodlett D.R., Aebersold R., and Hood L. Integrated genomic and proteomic analyses of a systematically perturbed metabolic network. Science 292 (2001) 929-934
36. Sontag E., Kiyatkin A., and Kholodenko B.N. Inferring dynamic architecture of cellular networks using time series of gene expression, protein and metabolite data. Bioinformatics 20 (2004) 1877-1886
37. Koza JR, Mydlowec W, Lanza G, Yu J, Keane MA: Automatic computational discovery of chemical reaction networks using genetic programming. Computational Discovery of Scientific Knowledge. Edited by Džeroski S, Todorovski L. Berlin: Springer; 2007:205-227. A genetic programming approach to finding biological networks, illustrated by reconstructing a part of a phospholipid cycle from simulated data.
38. Kikuchi S., Tominaga D., Arita M., Takahashi K., and Tomita M. Dynamic modeling of genetic networks using genetic algorithm and S-system. Bioinformatics 19 (2003) 643-650
39. Sontag E.D. For differential equations with r parameters, 2 r + 1 experiments are enough for identification. J Nonlinear Sci 12 (2002) 553-583
40. Brewer D., Barenco M., Callard R., Hubank M., and Stark J. Fitting ordinary differential equations to short time course data. Philos Trans A Math Phys Eng Sci 366 1865 (2008) 519-544. Introduces and evaluates a new efficient technique for estimating parameters in ordinary differential equations (linear in the parameters), particularly suited to situations where the number of data points is low.