Vitor V. Lopes

High‐throughput analysis of the plasmid bioproduction process in Escherichia coli by FTIR spectroscopy

Monitoring plasmid production systems is a lab intensive task. This article proposes a methodology based on FTIR spectroscopy and the use of chemometrics for the high‐throughput analysis of the plasmid bioproduction process in E. coli. For this study, five batch cultures with different initial medium compositions are designed to represent different biomass and plasmid production behavior, with the maximum plasmid and biomass concentrations varying from 11 to 95 mg L−1 and 6.8 to 12.8 g L−1, respectively, and the plasmid production per biomass varying from 0.4 to 5.1 mg g−1. After a short sample processing consisting of centrifugation and dehydration, the FTIR spectra are recorded from the collected cellular biomass using microtiter plates with 96 wells. After spectral pre‐processing, the predictive FTIR spectra models are derived by using partial least squares (PLS) regression with the wavenumber selection performed by a Monte‐Carlo strategy. Results show that it is possible to improve the PLS models by selecting specific spectral ranges. For the plasmid model, the spectral regions between 590–1,130, 1,670–2,025, and 2,565–3,280 cm−1 are found to be highly relevant. Whereas for the biomass, the best wavenumber selections are between 900–1,200, 1,500–1,800, and 2,850–3,200 cm−1. The optimized PLS models show a high coefficient of determination of 0.91 and 0.89 for the plasmid and biomass concentration, respectively. Additional PLS models for the prediction of the carbon sources glucose and glycerol and the by‐product acetic acid, based on metabolism‐induced correlations between the nutrients and the cellular biomass are also established. Biotechnol. Bioeng. 2012;109: 2279–2285.

On the use of Markov chain models for the analysis of wind power time-series

Wind energy is becoming a top contributor to the renewable energy mix, which raises potential reliability issues for the grid due to the fluctuating and intermittent nature of its source. This paper explores the use of Markov chain models for the analysis of wind power time-series. The proposed Markov chain model is based on a 2yr dataset collected from a wind turbine located in Portugal. The wind speed, direction and power variables are used to define the states and the transition matrix is determined using a maximum likelihood estimator based on multi-step transition data. The Markov chain model is analyzed by comparing the theoretically derived properties with their empirically determined analogues. Results show that the proposed model is capable of describing the observed statistics, such as wind speed and power probability density as well as the persistence statistics. It is demonstrated how the application of the Markov chain model can be used for the short-term prediction of wind power.

Impact of Weather Regimes on the Wind Power Ramp Forecast in Portugal

Short-term forecasting and diagnostic tools for severe changes of wind power production (power ramps) may provide reliable information for a secure power system operation at a small cost. Understanding the underlying role of the synoptic weather regimes (WRs) in triggering the wind power ramp events can be an added value to improve and complement the current forecast techniques. This work identifies and classifies the WRs over mainland Portugal associated with the occurrence of severe wind power ramps. The most representative WRs are identified on compressed surface level atmospheric data using principal component analysis by applying K-means clustering. The results show a strong association between some synoptic circulation patterns and step variations of the wind power production indicating the possibility to identify certain WRs that are prone to trigger severe wind power ramps, thus opening the possibility for future development of diagnostic warning systems for system operators use.

Parallelization Strategies for Spatial Agent-Based Models

Agent-based modeling (ABM) is a bottom-up modeling approach, where each entity of the system being modeled is uniquely represented as an independent decision-making agent. Large scale emergent behavior in ABMs is population sensitive. As such, the number of agents in a simulation should be able to reflect the reality of the system being modeled, which can be in the order of millions or billions of individuals in certain domains. A natural solution to reach acceptable scalability in commodity multi-core processors consists of decomposing models such that each component can be independently processed by a different thread in a concurrent manner. In this paper we present a multithreaded Java implementation of the PPHPC ABM, with two goals in mind: (1) compare the performance of this implementation with an existing NetLogo implementation; and, (2) study how different parallelization strategies impact simulation performance on a shared memory architecture. Results show that: (1) model parallelization can yield considerable performance gains; (2) distinct parallelization strategies offer specific trade-offs in terms of performance and simulation reproducibility; and, (3) PPHPC is a valid reference model for comparing distinct implementations or parallelization strategies, from both performance and statistical accuracy perspectives.

Uncovering wind turbine properties through two-dimensional stochastic modeling of wind dynamics

Using a method for stochastic data analysis borrowed from statistical physics, we analyze synthetic data from a Markov chain model that reproduces measurements of wind speed and power production in a wind park in Portugal. We show that our analysis retrieves indeed the power performance curve, which yields the relationship between wind speed and power production, and we discuss how this procedure can be extended for extracting unknown functional relationships between pairs of physical variables in general. We also show how specific features, such as the rated speed of the wind turbine or the descriptive wind speed statistics, can be related to the equations describing the evolution of power production and wind speed at single wind turbines.

Towards a standard model for research in agent-based modeling and simulation

This work was supported by the Fundacao para a Ciencia e a Tecnologia (FCT) projects UID/EEA/50009/2013, UID/MAT/04561/2013 and (P. RD0389) Incentivo/EEI/LA0009/2014, and partially funded with grant SFRH/BD/48310/2008, also from FCT. The author Vitor V. Lopes acknowledges the financial support from the Prometeo project of SENESCYT (Ecuador).

Model-independent comparison of simulation output

Abstract Computational models of complex systems are usually elaborate and sensitive to implementation details, characteristics which often affect their verification and validation. Model replication is a possible solution to this issue. It avoids biases associated with the language or toolkit used to develop the original model, not only promoting its verification and validation, but also fostering the credibility of the underlying conceptual model. However, different model implementations must be compared to assess their equivalence. The problem is, given two or more implementations of a stochastic model, how to prove that they display similar behavior? In this paper, we present a model comparison technique, which uses principal component analysis to convert simulation output into a set of linearly uncorrelated statistical measures, analyzable in a consistent, model-independent fashion. It is appropriate for ascertaining distributional equivalence of a model replication with its original implementation. Besides model-independence, this technique has three other desirable properties: a) it automatically selects output features that best explain implementation differences; b) it does not depend on the distributional properties of simulation output; and, c) it simplifies the modelers’ work, as it can be used directly on simulation outputs. The proposed technique is shown to produce similar results to the manual or empirical selection of output features when applied to a well-studied reference model.

Comparison of GA and PSO performance in parameter estimation of microbial growth models: A case-study using experimental data

This work examined the performance of a genetic algorithm (GA) and particle swarm optimization (PSO) in parameter estimation for a yeast growth kinetic model. Fitting the models predictions simultaneously to three replicates of the same experiment, we used the variability among replicates as a criterion to evaluate the optimization result, since it reflects the biological variability characteristic of these systems. The performance of each algorithm was studied using 12 distinct tuning settings: a) in the GA, the tuning addressed different combinations of crossover fraction, and crossover and mutation functions; b) in the PSO, three different convergence behavior types (convergent with and without oscillations and divergent) were tested and the local and global weights were varied. The best objective function values were obtained when the PSO had convergent oscillatory behavior and a local acceleration larger than the global acceleration.

Simulating antigenic drift and shift in influenza A

Computational models of the immune system and pathogenic agents have several applications, such as theory testing and validation, or as a complement to first stages of drug trials. One possible application is the prediction of the lethality of new Influenza A strains, which are constantly created due to antigenic drift and shift. Here, we present an agent-based model of immune-influenza A dynamics, with focus on low level molecular antigen-antibody interactions, in order to study antigenic drift and shift events, and analyze the virulence of emergent strains. At this stage of the investigation, results are presented and discussed from a qualitative point of view against recent and generally recognized immunology and influenza literature.

A template model for agent-based simulations

Agent-based modeling (ABM) is a bottom-up modeling approach, where each entity of the system being modeled is uniquely represented as an independent decision-making agent. ABMs are very sensitive to implementation details. Thus, it is very easy to inadvertently introduce changes which modify model dynamics. Such problems usually arise due to the lack of transparency in model descriptions, which constrains how models are assessed, implemented and replicated. In this paper, we present a template ABM which aims to serve as a basis for a series of investigations, including, but not limited to, conceptual model specification, statistical analysis of simulation output, model comparison and model parallelization. This paper focuses on the first two aspects (conceptual model specification and statistical analysis of simulation output), also providing a canonical implementation of the template ABM, such that it serves as a complete reference to the presented model. Additionally, this study is presented in a tutorial fashion, and can be used as a road map for simulation practitioners who wish to improve the way they communicate their ABMs.