Finding Metabolic Pathways in Decision Forests

Conference paper
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)


Data of metabolite concentrations in tissue samples is a powerful source of information about metabolic activity in organisms. Several methods have already been reported that allow for inferences about genetic regulatory networks using expression profiling data. Here we adapted these techniques for metabolic concentration data. While somewhat accurate in predicting certain properties these methods encounter problems when dealing with networks containing a high number of vertices (τ 50). To circumvent these difficulties, larger data sets are usually reduced a priori by means of preprocessing techniques. Our approach allows to make network inferences using ensembles of decision trees that can handle almost any amount of vertices, and thus to avoid time consuming preprocessing steps. The technique works on a bootstrap principle and heuristically searches for partially correlated relations between all objects. We applied this approach to synthetically generated data as well as on data taken from real experiments.


Decision Tree Partial Correlation Target Variable Target Attribute Genetic Regulatory Network 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. FIEHN, O., KOPKA, J., DÖRMANN, P., ALTMANN, T., TRETHEWEY, R.N., and WILMITZER, L. (2000): Metabolite profiling for plant functional genomics. In: Proc. Natl. Acad. Sci. USA, 95, 14863–14868.Google Scholar
  2. HORIMOTO, K. and TOH, H. (2001): Automatic System for Inferring a Network from Gene Expression Profiles. Genome Informatics, 12, 270–271.Google Scholar
  3. KOSE, F., WECKWETH, W., LINKE, T., and FIEHN, O. (2001): Visualising plant metabolomic correlation networks using clique-metabolite matrices. Bioinformatics, 17, 1198–1208.CrossRefGoogle Scholar
  4. MITCHELL, T. (1997): Machine Learning. McGraw-Hill, Boston, USA.Google Scholar
  5. PE’ER, D, REGEV, A, ELIDAN, G., and FFRIEDMAN, N. (2001): Inferring subnetworks from perturbed expression profiles. Bioinformatics, 17,Suppl. 1, 215–S224.Google Scholar
  6. SHIPLEY, B. (2002): Cause and Correlation in Biology. Cambridge University Press, Cambridge, UK.Google Scholar
  7. SILVERMAN, B.W. (1981): Using Kernel Density Estimates to investigate Multimodality. Journal of Royal Statistical Society, Series B, 43, 97–99.Google Scholar
  8. STATISTICA (2002): Electronic Textbook StatSoft. Scholar
  9. QUINLAN, J.R. (1993): C4.5: Programs for Machine Learning. Morgan Kaufmann, San Mateo, USA.Google Scholar

Copyright information

© Springer-Verlag Berlin · Heidelberg 2005

Authors and Affiliations

  1. 1.Institut für InformatikUniversität PotsdamPotsdamGermany
  2. 2.Max-Planck-Institut für molekulare PflanzenphysiologiePotsdamGermany

Personalised recommendations