Ignacio Ponzoni

On Stopping Criteria for Genetic Algorithms

In this work we present a critical analysis of various aspects associated with the specification of termination conditions for simple genetic algorithms. The study, which is based on the use of Markov chains, identifies the main difficulties that arise when one wishes to set meaningful upper bounds for the number of iterations required to guarantee the convergence of such algorithms with a given confidence level. The latest trends in the design of stopping rules for evolutionary algorithms in general are also put forward and some proposals to overcome existing limitations in this respect are suggested.

CGD-GA: A graph-based genetic algorithm for sensor network design

Abstract The foundations and implementation of a genetic algorithm (GA) for instrumentation purposes are presented in this paper. The GA constitutes an initialization module of a decision support system for sensor network design. The method development entailed the definition of the individual’s representation as well as the design of a graph-based fitness function, along with the formulation of several other ad hoc implemented features. The performance and effectiveness of the GA were assessed by initializing the instrumentation design of an ammonia synthesis plant. The initialization provided by the GA succeeded in accelerating the sensor network design procedures. It also accomplished a great improvement in the overall quality of the resulting instrument configuration. Therefore, the GA constitutes a valuable tool for the treatment of real industrial problems.

Pathway network inference from gene expression data

BackgroundThe development of high-throughput omics technologies enabled genome-wide measurements of the activity of cellular elements and provides the analytical resources for the progress of the Systems Biology discipline. Analysis and interpretation of gene expression data has evolved from the gene to the pathway and interaction level, i.e. from the detection of differentially expressed genes, to the establishment of gene interaction networks and the identification of enriched functional categories. Still, the understanding of biological systems requires a further level of analysis that addresses the characterization of the interaction between functional modules.ResultsWe present a novel computational methodology to study the functional interconnections among the molecular elements of a biological system. The PANA approach uses high-throughput genomics measurements and a functional annotation scheme to extract an activity profile from each functional block -or pathway- followed by machine-learning methods to infer the relationships between these functional profiles. The result is a global, interconnected network of pathways that represents the functional cross-talk within the molecular system. We have applied this approach to describe the functional transcriptional connections during the yeast cell cycle and to identify pathways that change their connectivity in a disease condition using an Alzheimer example.ConclusionsPANA is a useful tool to deepen in our understanding of the functional interdependences that operate within complex biological systems. We show the approach is algorithmically consistent and the inferred network is well supported by the available functional data. The method allows the dissection of the molecular basis of the functional connections and we describe the different regulatory mechanisms that explain the networks topology obtained for the yeast cell cycle data.

Target-Driven Subspace Mapping Methods and Their Applicability Domain Estimation.

This work describes a methodology for assisting virtual screening of drugs during the early stages of the drug development process. This methodology is proposed to improve the reliability of in silico property prediction and it is structured in two steps. Firstly, a transformation is sought for mapping a high‐dimensional space defined by potentially redundant or irrelevant molecular descriptors into a low‐dimensional application‐related space. For this task we evaluate three different target‐driven subspace mapping methods, out of which we highlight the recent Correlative Matrix Mapping (CMM) as the most stable. Secondly, we apply an applicability domain model on the low‐dimensional space for assessing confidentiality of compound classification. By a probabilistic framework the applicability domain approach identifies poorly represented compounds in the training set (extrapolation problems) and regions in the space where the uncertainty about the correct class is higher than normal (interpolation problems). This two‐step approach represents an important contribution to the development of confident prediction tools in the chemoinformatics area, where the field is in need of both interpretable models and methods that estimate the confidence of predictions.

Microarray Biclustering: A Novel Memetic Approach Based on the PISA Platform

In this paper, a new memetic approach that integrates a Multi-Objective Evolutionary Algorithm (MOEA) with local search for microarray biclustering is presented. The original features of this proposal are the consideration of opposite regulation and incorporation of a mechanism for tuning the balance between the size and row variance of the biclusters. The approach was developed according to the Platform and Programming Language Independent Interface for Search Algorithms (PISA) framework, thus achieving the possibility of testing and comparing several different memetic MOEAs. The performance of the MOEA strategy based on the SPEA2 performed better, and its resulting biclusters were compared with those obtained by a multi-objective approach recently published. The benchmarks were two datasets corresponding to Saccharomyces cerevisiae and human B-cells Lymphoma . Our proposal achieves a better proportion of coverage of the gene expression data matrix, and it also obtains biclusters with new features that the former existing evolutionary strategies can not detect.

BiHEA: A Hybrid Evolutionary Approach for Microarray Biclustering

In this paper a new hybrid approach that integrates an evolutionary algorithm with local search for microarray biclustering is presented. The novelty of this proposal is constituted by the incorporation of two mechanisms: the first one avoids loss of good solutions through generations and overcomes the high degree of overlap in the final population; and the other one preserves an adequate level of genotypic diversity. The performance of the memetic strategy was compared with the results of several salient biclustering algorithms over synthetic data with different overlap degrees and noise levels. In this regard, our proposal achieves results that outperform the ones obtained by the referential methods. Finally, a study on real data was performed in order to demonstrate the biological relevance of the results of our approach.

Discretization of gene expression data revised

Gene expression measurements represent the most important source of biological data used to unveil the interaction and functionality of genes. In this regard, several data mining and machine learning algorithms have been proposed that require, in a number of cases, some kind of data discretization to perform the inference. Selection of an appropriate discretization process has a major impact on the design and outcome of the inference algorithms, as there are a number of relevant issues that need to be considered. This study presents a revision of the current state-of-the-art discretization techniques, together with the key subjects that need to be considered when designing or selecting a discretization approach for gene expression data.

Discovering time-lagged rules from microarray data using gene profile classifiers.

BackgroundGene regulatory networks have an essential role in every process of life. In this regard, the amount of genome-wide time series data is becoming increasingly available, providing the opportunity to discover the time-delayed gene regulatory networks that govern the majority of these molecular processes.ResultsThis paper aims at reconstructing gene regulatory networks from multiple genome-wide microarray time series datasets. In this sense, a new model-free algorithm called GRNCOP2 (G ene R egulatory N etwork inference by C ombinatorial OP timization 2), which is a significant evolution of the GRNCOP algorithm, was developed using combinatorial optimization of gene profile classifiers. The method is capable of inferring potential time-delay relationships with any span of time between genes from various time series datasets given as input. The proposed algorithm was applied to time series data composed of twenty yeast genes that are highly relevant for the cell-cycle study, and the results were compared against several related approaches. The outcomes have shown that GRNCOP2 outperforms the contrasted methods in terms of the proposed metrics, and that the results are consistent with previous biological knowledge. Additionally, a genome-wide study on multiple publicly available time series data was performed. In this case, the experimentation has exhibited the soundness and scalability of the new method which inferred highly-related statistically-significant gene associations.ConclusionsA novel method for inferring time-delayed gene regulatory networks from genome-wide time series datasets is proposed in this paper. The method was carefully validated with several publicly available data sets. The results have demonstrated that the algorithm constitutes a usable model-free approach capable of predicting meaningful relationships between genes, revealing the time-trends of gene regulation.

Visual analytics in cheminformatics: user-supervised descriptor selection for QSAR methods

BackgroundThe design of QSAR/QSPR models is a challenging problem, where the selection of the most relevant descriptors constitutes a key step of the process. Several feature selection methods that address this step are concentrated on statistical associations among descriptors and target properties, whereas the chemical knowledge is left out of the analysis. For this reason, the interpretability and generality of the QSAR/QSPR models obtained by these feature selection methods are drastically affected. Therefore, an approach for integrating domain expert’s knowledge in the selection process is needed for increase the confidence in the final set of descriptors.ResultsIn this paper a software tool, which we named Visual and Interactive DEscriptor ANalysis (VIDEAN), that combines statistical methods with interactive visualizations for choosing a set of descriptors for predicting a target property is proposed. Domain expertise can be added to the feature selection process by means of an interactive visual exploration of data, and aided by statistical tools and metrics based on information theory. Coordinated visual representations are presented for capturing different relationships and interactions among descriptors, target properties and candidate subsets of descriptors. The competencies of the proposed software were assessed through different scenarios. These scenarios reveal how an expert can use this tool to choose one subset of descriptors from a group of candidate subsets or how to modify existing descriptor subsets and even incorporate new descriptors according to his or her own knowledge of the target property.ConclusionsThe reported experiences showed the suitability of our software for selecting sets of descriptors with low cardinality, high interpretability, low redundancy and high statistical performance in a visual exploratory way. Therefore, it is possible to conclude that the resulting tool allows the integration of a chemist’s expertise in the descriptor selection process with a low cognitive effort in contrast with the alternative of using an ad-hoc manual analysis of the selected descriptors.

ModGen: a model generator for instrumentation analysis

Abstract A computer software tool for the automatic generation of steady-state process models to be used in instrumentation analysis was developed. We describe the program, called ModGen, discussing its main advantages and potential benefits. ModGen constitutes the front-end of a complete decision support system (DSS) for plant instrumentation design and revamp. This DSS is currently under development. The paper concludes with the description of ModGens application to the classification of unmeasured variables of an existing medium-size process plant by means of GS-FLCNs structural technique for observability analysis.