Miguel Rocha

OptFlux: an open-source software platform for in silico metabolic engineering

BackgroundOver the last few years a number of methods have been proposed for the phenotype simulation of microorganisms under different environmental and genetic conditions. These have been used as the basis to support the discovery of successful genetic modifications of the microbial metabolism to address industrial goals. However, the use of these methods has been restricted to bioinformaticians or other expert researchers. The main aim of this work is, therefore, to provide a user-friendly computational tool for Metabolic Engineering applications.ResultsOptFlux is an open-source and modular software aimed at being the reference computational application in the field. It is the first tool to incorporate strain optimization tasks, i.e., the identification of Metabolic Engineering targets, using Evolutionary Algorithms/Simulated Annealing metaheuristics or the previously proposed OptKnock algorithm. It also allows the use of stoichiometric metabolic models for (i) phenotype simulation of both wild-type and mutant organisms, using the methods of Flux Balance Analysis, Minimization of Metabolic Adjustment or Regulatory on/off Minimization of Metabolic flux changes, (ii) Metabolic Flux Analysis, computing the admissible flux space given a set of measured fluxes, and (iii) pathway analysis through the calculation of Elementary Flux Modes.OptFlux also contemplates several methods for model simplification and other pre-processing operations aimed at reducing the search space for optimization algorithms.The software supports importing/exporting to several flat file formats and it is compatible with the SBML standard. OptFlux has a visualization module that allows the analysis of the model structure that is compatible with the layout information of Cell Designer, allowing the superimposition of simulation results with the model graph.ConclusionsThe OptFlux software is freely available, together with documentation and other resources, thus bridging the gap from research in strain optimization algorithms and the final users. It is a valuable platform for researchers in the field that have available a number of useful tools. Its open-source nature invites contributions by all those interested in making their methods available for the community.Given its plug-in based architecture it can be extended with new functionalities. Currently, several plug-ins are being developed, including network topology analysis tools and the integration with Boolean network based regulatory models.

Particle swarms for feedforward neural network training

Particle swarm is an optimization paradigm for real-valued functions, based on the social dynamics of group interaction. We propose its application to the training of neural networks. Comparative tests were carried out, for classification and regression tasks.

Modeling formalisms in Systems Biology

Systems Biology has taken advantage of computational tools and high-throughput experimental data to model several biological processes. These include signaling, gene regulatory, and metabolic networks. However, most of these models are specific to each kind of network. Their interconnection demands a whole-cell modeling framework for a complete understanding of cellular systems. We describe the features required by an integrated framework for modeling, analyzing and simulating biological processes, and review several modeling formalisms that have been used in Systems Biology including Boolean networks, Bayesian networks, Petri nets, process algebras, constraint-based models, differential equations, rule-based models, interacting state machines, cellular automata, and agent-based models. We compare the features provided by different formalisms, and discuss recent approaches in the integration of these formalisms, as well as possible directions for the future.

Natural computation meta-heuristics for the in silico optimization of microbial strains

BackgroundOne of the greatest challenges in Metabolic Engineering is to develop quantitative models and algorithms to identify a set of genetic manipulations that will result in a microbial strain with a desirable metabolic phenotype which typically means having a high yield/productivity. This challenge is not only due to the inherent complexity of the metabolic and regulatory networks, but also to the lack of appropriate modelling and optimization tools. To this end, Evolutionary Algorithms (EAs) have been proposed for in silico metabolic engineering, for example, to identify sets of gene deletions towards maximization of a desired physiological objective function. In this approach, each mutant strain is evaluated by resorting to the simulation of its phenotype using the Flux-Balance Analysis (FBA) approach, together with the premise that microorganisms have maximized their growth along natural evolution.ResultsThis work reports on improved EAs, as well as novel Simulated Annealing (SA) algorithms to address the task of in silico metabolic engineering. Both approaches use a variable size set-based representation, thereby allowing the automatic finding of the best number of gene deletions necessary for achieving a given productivity goal. The work presents extensive computational experiments, involving four case studies that consider the production of succinic and lactic acid as the targets, by using S. cerevisiae and E. coli as model organisms. The proposed algorithms are able to reach optimal/near-optimal solutions regarding the production of the desired compounds and presenting low variability among the several runs.ConclusionThe results show that the proposed SA and EA both perform well in the optimization task. A comparison between them is favourable to the SA in terms of consistency in obtaining optimal solutions and faster convergence. In both cases, the use of variable size representations allows the automatic discovery of the approximate number of gene deletions, without compromising the optimality of the solutions.

Preventing premature convergence to local optima in genetic algorithms via random offspring generation

The Genetic Algorithms (GAs) paradigm is being used increasingly in search and optimization problems. The method has shown to be efficient and robust in a considerable number of scientific domains, where the complexity and cardinality of the problems considered elected themselves as key factors to be taken into account. However, there are still some insufficiencies; indeed, one of the major problems usually associated with the use of GAs is the premature convergence to solutions coding local optima of the objective function. The problem is tightly related with the loss of genetic diversity of the GA’s population, being the cause of a decrease on the quality of the solutions found. Out of question, this fact has lead to the development of different techniques aiming to solve, or at least to minimize the problem; traditional methods usually work to maintain a certain degree of genetic diversity on the target populations, without affecting the convergence process of the GA. In one’s work, some of these techniques are compared and an innovative one, the Random Offspring Generation, is presented and evaluated in its merits. The Traveling Salesman Problem is used as a benchmark.

Evolution of neural networks for classification and regression

Although Artificial Neural Networks (ANNs) are important data mining techniques, the search for the optimal ANN is a challenging task: the ANN should learn the input-output mapping without overfitting the data and training algorithms may get trapped in local minima. The use of Evolutionary Computation (EC) is a promising alternative for ANN optimization. This work presents two hybrid EC/ANN algorithms: the first evolves neural topologies while the latter performs simultaneous optimization of architectures and weights. Sixteen real-world tasks were used to test these strategies. Competitive results were achieved when compared with a heuristic model selection and other Data Mining algorithms.

Evolving Time Series Forecasting ARMA Models

Time Series Forecasting (TSF) allows the modeling of complex systems as “black-boxes”, being a focus of attention in several research arenas such as Operational Research, Statistics or Computer Science. Alternative TSF approaches emerged from the Artificial Intelligence arena, where optimization algorithms inspired on natural selection processes, such as Evolutionary Algorithms (EAs), are popular. The present work reports on a two-level architecture, where a (meta-level) binary EA will search for the best ARMA model, being the parameters optimized by a (low-level) EA, which encodes real values. The handicap of this approach is compared with conventional forecasting methods, being competitive.

Bridging the layers: towards integration of signal transduction, regulation and metabolism into mathematical models

Mathematical modelling is increasingly becoming an indispensable tool for the study of cellular processes, allowing their analysis in a systematic and comprehensive manner. In the vast majority of the cases, models focus on specific subsystems, and in particular describe either metabolism, gene expression or signal transduction. Integrated models that are able to span and interconnect these layers are, by contrast, rare as their construction and analysis face multiple challenges. Such methods, however, would represent extremely useful tools to understand cell behaviour, with application in distinct fields of biological and medical research. In particular, they could be useful tools to study genotype-phenotype mappings, and the way they are affected by specific conditions or perturbations. Here, we review existing computational approaches that integrate signalling, gene regulation and/or metabolism. We describe existing challenges, available methods and point at potentially useful strategies.

Internet Traffic Forecasting using Neural Networks

The forecast of Internet traffic is an important issue that has received few attention from the computer networks field. By improving this task, efficient traffic engineering and anomaly detection tools can be created, resulting in economic gains from better resource management. This paper presents a neural network ensemble (NNE) for the prediction of TCP/IP traffic using a time series forecasting (TSF) point of view. Several experiments were devised by considering real-world data from two large Internet Service Providers. In addition, different time scales (e.g. every five minutes and hourly) and forecasting horizons were analyzed. Overall, the NNE approach is competitive when compared with other TSF methods (e.g. Holt-Winters and ARIMA).

Metabolomics combined with chemometric tools (PCA, HCA, PLS-DA and SVM) for screening cassava (Manihot esculenta Crantz) roots during postharvest physiological deterioration

Cassava roots are an important source of dietary and industrial carbohydrates and suffer markedly from postharvest physiological deterioration (PPD). This paper deals with metabolomics combined with chemometric tools for screening the chemical and enzymatic composition in several genotypes of cassava roots during PPD. Metabolome analyses showed increases in carotenoids, flavonoids, anthocyanins, phenolics, reactive scavenging species, and enzymes (superoxide dismutase family, hydrogen peroxide, and catalase) until 3-5days postharvest. PPD correlated negatively with phenolics and carotenoids and positively with anthocyanins and flavonoids. Chemometric tools such as principal component analysis, partial least squares discriminant analysis, and support vector machines discriminated well cassava samples and enabled a good prediction of samples. Hierarchical clustering analyses grouped samples according to their levels of PPD and chemical compositions.