Software for dynamic analysis of tracer-based metabolomic data: estimation of metabolic fluxes and their statistical analysis
| Autor: | Paul W. N. Lee, Vitaly A. Selivanov, Marta Cascante, Silvia Marin |
|---|---|
| Rok vydání: | 2006 |
| Předmět: |
Statistics and Probability
Computer science Monte Carlo method Models Biological Biochemistry Gas Chromatography-Mass Spectrometry Metabolism regulation Isotopomers Metabolomics Metabolic flux analysis Statistics Range (statistics) Animals Humans Molecular Biology Carbon Isotopes Models Statistical Isotope Computational Biology Experimental data Rats Computer Science Applications Computational Mathematics Metabolism Liver Computational Theory and Mathematics Isotope Labeling Simulated annealing Programming Languages Biological system Monte Carlo Method Algorithms Software |
| Zdroj: | Bioinformatics. 22:2806-2812 |
| ISSN: | 1367-4811 1367-4803 |
| DOI: | 10.1093/bioinformatics/btl484 |
| Popis: | Motivation: Metabolic flux analysis of biochemical reaction networks using isotope tracers requires software tools that can analyze the dynamics of isotopic isomer (isotopomer) accumulation in metabolites and reveal the underlying kinetic mechanisms of metabolism regulation. Since existing tools are restricted by the isotopic steady state and remain disconnected from the underlying kinetic mechanisms, we have recently developed a novel approach for the analysis of tracer-based metabolomic data that meets these requirements. The present contribution describes the last step of this development: implementation of (i) the algorithms for the determination of the kinetic parameters and respective metabolic fluxes consistent with the experimental data and (ii) statistical analysis of both fluxes and parameters, thereby lending it a practical application. Results: The C++ applications package for dynamic isotopomer distribution data analysis was supplemented by (i) five distinct methods for resolving a large system of differential equations; (ii) the ‘simulated annealing’ algorithm adopted to estimate the set of parameters and metabolic fluxes, which corresponds to the global minimum of the difference between the computed and measured isotopomer distributions; and (iii) the algorithms for statistical analysis of the estimated parameters and fluxes, which use the covariance matrix evaluation, as well as Monte Carlo simulations. An example of using this tool for the analysis of 13C distribution in the metabolites of glucose degradation pathways has demonstrated the evaluation of optimal set of parameters and fluxes consistent with the experimental pattern, their range and statistical significance, and also the advantages of using dynamic rather than the usual steady-state method of analysis. Availability: Software is available free from Contact: martacascante@ub.edu |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|