Kuffner, R., Zimmer, R., and Lengauer, T.. "Pathway analysis in metabolic databases via differential metabolic display (DMD)." Bioinformatics. 16
(9).
2000.
pp. 825-36.
[ .pdf ] [ PubMed ] [ WebSite ]
MOTIVATION: A number of metabolic databases are available electronically, some with features for querying and visualizing metabolic pathways and regulatory networks. We present a unifying, systematic approach based on PETRI nets for storing, displaying, comparing, searching and simulating such nets from a number of different sources. RESULTS: Information from each data source is extracted and compiled into a PETRI net. Such PETRI nets then allow to investigate the (differential) content in metabolic databases, to map and integrate genomic information and functional annotations, to compare sequence and metabolic databases with respect to their functional annotations, and to define, generate and search paths and pathways in nets. We present an algorithm to systematically generate all pathways satisfying additional constraints in such PETRI nets. Finally, based on the set of valid pathways, so-called differential metabolic displays (DMDs) are introduced to exhibit specific differences between biological systems, i.e. different developmental states, disease states, or different organisms, on the level of paths and pathways. DMDs will be useful for target finding and function prediction, especially in the context of the interpretation of expression data.
Keywords: *Algorithms ; Catalysis ; Computational Biology_*methods ; Computer Simulation ; *Data Display ; *Databases Factual ; Enzymes_genetics ; Enzymes_metabolism ; Glycolysis ; Metabolism_*physiology ; Mycoplasma_metabolism ; Yeasts_metabolism
Mavrovouniotis, M.L.. "Identification of localized and distributed bottlenecks in metabolic pathways." Proc Int Conf Intell Syst Mol Biol.
vol. 1.
1993.
pp. 275-83.
[ PubMed ]
The usual thermodynamic evaluation, based solely on the Standard Gibbs Energy of reaction, does not take into account the permissible ranges of concentrations of metabolites, and it faces further difficulties when, instead of isolated reactions, we are examining whole pathways. For pathways, we seek not only to decide whether they are feasible but also to pinpoint the pathway segment that causes any thermodynamic difficulties. We define a set of scaled quantities which reformulate the thermodynamic-feasibility problem for the whole pathway. We present an algorithm which analyzes individual reactions and selective construction of larger subpathways and uncovers localized and distributed thermodynamic bottlenecks of the biotransformation. This type of thermodynamic treatment contributes to the effort to include more physical, chemical, and biological factors in the computer-aided analysis of metabolic pathways.
Keywords: *Algorithms ; *Computer Simulation ; Glycolysis ; *Metabolism ; *Models Biological ; *Thermodynamics
Rana, O.F.. "Automating parallel implementation of neural learning algorithms." Int J Neural Syst. 10
(3).
2000.
pp. 227-41.
[ PubMed ]
Neural learning algorithms generally involve a number of identical processing units, which are fully or partially connected, and involve an update function, such as a ramp, a sigmoid or a Gaussian function for instance. Some variations also exist, where units can be heterogeneous, or where an alternative update technique is employed, such as a pulse stream generator. Associated with connections are numerical values that must be adjusted using a learning rule, and and dictated by parameters that are learning rule specific, such as momentum, a learning rate, a temperature, amongst others. Usually, neural learning algorithms involve local updates, and a global interaction between units is often discouraged, except in instances where units are fully connected, or involve synchronous updates. In all of these instances, concurrency within a neural algorithm cannot be fully exploited without a suitable implementation strategy. A design scheme is described for translating a neural learning algorithm from inception to implementation on a parallel machine using PVM or MPI libraries, or onto programmable logic such as FPGAs. A designer must first describe the algorithm using a specialised Neural Language, from which a Petri net (PN) model is constructed automatically for verification, and building a performance model. The PN model can be used to study issues such as synchronisation points, resource sharing and concurrency within a learning rule. Specialised constructs are provided to enable a designer to express various aspects of a learning rule, such as the number and connectivity of neural nodes, the interconnection strategies, and information flows required by the learning algorithm. A scheduling and mapping strategy is then used to translate this PN model onto a multiprocessor template. We demonstrate our technique using a Kohonen and backpropagation learning rules, implemented on a loosely coupled workstation cluster, and a dedicated parallel machine, with PVM libraries.
Keywords: *Algorithms ; Artificial Intelligence ; Computers ; Models Neurological ; *Neural Networks (Computer) ; Programming Languages
Schuster, S., Pfeiffer, T., Moldenhauer, F., Koch, I., and Dandekar, T.. "Exploring the pathway structure of metabolism: decomposition into subnetworks and application to Mycoplasma pneumoniae." Bioinformatics. 18
(2).
2002.
pp. 351-61.
[ PubMed ] [ WebSite ]
MOTIVATION: Reconstructing and analyzing the metabolic map of microorganisms is an important challenge in bioinformatics. Pathway analysis of large metabolic networks meets with the problem of combinatorial explosion of pathways. Therefore, appropriate algorithms for an automated decomposition of these networks into smaller subsystems are needed. RESULTS: A decomposition algorithm for metabolic networks based on the local connectivity of metabolites is presented. Interrelations of this algorithm with alternative methods proposed in the literature and the theory of small world networks are discussed. The applicability of our method is illustrated by an analysis of the metabolism of Mycoplasma pneumoniae, which is an organism of considerable medical interest. The decomposition gives rise to 19 subnetworks. Three of these are here discussed in biochemical terms: arginine degradation, the tetrahydrofolate system, and nucleotide metabolism. The interrelations of pathway analysis of biochemical networks with Petri net theory are outlined.
Keywords: Algorithms ; Arginine_metabolism ; Computational Biology ; *Metabolism ; Models Biological ; Mycoplasma pneumoniae_*metabolism ; Nucleotides_metabolism ; *Software