Carlos-Andrés Peña-Reyes - Profile on Academia.edu (original) (raw)
Papers by Carlos-Andrés Peña-Reyes
Stability of Feature Selection Methods: A Study of Metrics Across Different Gene Expression Datasets
Bioinformatics and Biomedical Engineering, 2020
Analysis of gene-expression data often requires that a gene (feature) subset is selected and many... more Analysis of gene-expression data often requires that a gene (feature) subset is selected and many feature selection (FS) methods have been devised. However, FS methods often generate different lists of features for the same dataset and users then have to choose which list to use. One approach to support this choice is to apply stability metrics on the generated lists and selecting lists on that base. The aim of this study is to investigate the behavior of stability metrics applied to feature subsets generated by FS methods. The experiments in this work explore a plethora of gene expression datasets, FS methods, and expected number of features to compare several stability metrics. The stability metrics have been used to compare five feature selection methods (SVM, SAM, ReliefF, RFE + RF and LIMMA) on gene expression datasets from the EBI repository. Results show that the studied stability metrics display a high amount of variability. The reason behind this is not clear yet and is being further investigated. The final objective of the research, that is to define how to select a FS method, is an ongoing work whose partial findings are reported herein.
Bioinformatics and Biomedical Engineering, 2017
The emergence and rapid dissemination of antibiotic resistance, worldwide, threatens medical prog... more The emergence and rapid dissemination of antibiotic resistance, worldwide, threatens medical progress and calls for innovative approaches for the management of multidrug resistant infections. Phage-therapy, i.e., the use of viruses (phages) that specifically infect and kill bacteria during their life cycle, is a re-emerging and promising alternative to solve this problem. The success of phage therapy mainly relies on the exact matching between the target pathogenic bacteria and the therapeutic phage. Currently, there are only a few tools or methodologies that efficiently predict phage-bacteria interactions suitable for the phage therapy, and the pairs phage-bacterium are thus empirically tested in laboratory. In this paper we present an original methodology, based on an ensemble-learning approach, to predict whether or not a given pair of phage-bacteria would interact. Using publicly available information from Genbank and phagesdb.org, we assembled a dataset containing more than two thousand phage-bacterium interactions with their corresponding genomes. A set of informative features, extracted from these genomes, form the base of the quantitative datasets used to train our predictive models. These features include the distribution of predicted protein-protein interaction scores, as well as the amino acid frequency, the chemical composition, and the molecular weight of such proteins. Using an independent test dataset to evaluate the performance of our methodology, our approach gets encouraging performance with more than 90% of accuracy, specificity, and sensitivity.
Agent-based modeling of mesenchymal stem cells on a 3D-printed bio-device for the regenerative treatment of the infarcted myocardium
A myocardial infarction (MI) implies an obstruction of the blood flow due to a blockade in the co... more A myocardial infarction (MI) implies an obstruction of the blood flow due to a blockade in the coronary artery, which results in the cellular necrosis of the myocardium due to ischemia. Mesenchymal stem cells (MSCs) are non-differentiated cells with the ability to differentiate into any cell type. A regenerative treatment has been proposed for the regeneration of the infarcted myocardium, as the existing ones do not aim at the regeneration of the necrosed tissue. The treatment proposed is based on the design and construction of a 3D-printed biodevice, containing MSCs and other cell types, which can be later implanted on the patient's myocardium.A good understanding of the behavior of the cells in the biodevice is important to guide the treatment research. Thereby, the development of a computational model of the behavior of the MSC and other cell types used in the 3D-printed biodevice is carried out, allowing to better understand its dynamics and the effect that different variabl...
This paper describes a genetic-fuzzy approach for controlling a nonlinear model of the HIV immuno... more This paper describes a genetic-fuzzy approach for controlling a nonlinear model of the HIV immunology. The approach is set up to find mamdani fuzzy controllers capable of boosting the immune response while reducing the systemic cost to the body due to the use of high efficacy drugs. General aspects of the genetic fuzzy system are described while special emphasis is given to control results. The best controller found from several evolutionary runs is analyzed in terms of control performance and it is compared with a traditional medical procedure. Simulation results show that the fuzzy control approach provides a better long-term treatment.
Data, 2019
Regenerative medicine involves methods to control and modify normal tissue repair processes. Poly... more Regenerative medicine involves methods to control and modify normal tissue repair processes. Polymer and cell constructs are under research to create tissue that replaces the affected area in cardiac tissue after myocardial infarction (MI). The aim of the present study is to evaluate the behavior of differentiated and undifferentiated mesenchymal stem cells (MSCs) in vitro and in silico and to compare the results that both offer when it comes to the design process of biodevices for the treatment of infarcted myocardium in biomodels. To assess in vitro behavior, MSCs are isolated from rat bone marrow and seeded undifferentiated and differentiated in multiple scaffolds of a gelled biomaterial. Subsequently, cell behavior is evaluated by trypan blue and fluorescence microscopy, which showed that the cells presented high viability and low cell migration in the biomaterial. An agent-based model intended to reproduce as closely as possible the behavior of individual MSCs by simulating cel...
BMC Bioinformatics, 2018
Background: Antibiotic resistance and its rapid dissemination around the world threaten the effic... more Background: Antibiotic resistance and its rapid dissemination around the world threaten the efficacy of currently-used medical treatments and call for novel, innovative approaches to manage multi-drug resistant infections. Phage therapy, i.e., the use of viruses (phages) to specifically infect and kill bacteria during their life cycle, is one of the most promising alternatives to antibiotics. It is based on the correct matching between a target pathogenic bacteria and the therapeutic phage. Nevertheless, correctly matching them is a major challenge. Currently, there is no systematic method to efficiently predict whether phage-bacterium interactions exist and these pairs must be empirically tested in laboratory. Herein, we present our approach for developing a computational model able to predict whether a given phage-bacterium pair can interact based on their genome. Results: Based on public data from GenBank and phagesDB.org, we collected more than a thousand positive phage-bacterium interactions with their complete genomes. In addition, we generated putative negative (i.e., non-interacting) pairs. We extracted, from the collected genomes, a set of informative features based on the distribution of predictive protein-protein interactions and on their primary structure (e.g. amino-acid frequency, molecular weight and chemical composition of each protein). With these features, we generated multiple candidate datasets to train our algorithms. On this base, we built predictive models exhibiting predictive performance of around 90% in terms of F1-score, sensitivity, specificity, and accuracy, obtained on the test set with 10-fold cross-validation. Conclusion: These promising results reinforce the hypothesis that machine learning techniques may produce highly-predictive models accelerating the search of interacting phage-bacteria pairs.
![Research paper thumbnail of Implementing Interval Type-2 Fuzzy Processors [Developmental Tools]](https://attachments.academia-assets.com/85062337/thumbnails/1.jpg)
](IEEE Computational Intelligence Magazine, 2007
2006 IEEE International Conference on Fuzzy Systems, 2006
This paper presents a genetic fuzzy system approach to control a nonlinear dynamic model of the H... more This paper presents a genetic fuzzy system approach to control a nonlinear dynamic model of the HIV infection. The system is conceived to find mamdani fuzzy controllers that are capable of boosting the immune response while reducing the impact on the body because of the use of potentially toxic medicaments. General aspects of the used approach are described while special emphasis is given to the evolutionary mechanism. The best solution found from several evolutionary runs is analyzed in terms of control and interpretability issues.
Exploration of multiclass and one-class learning methods for prediction of phage-bacteria interaction at strain level
2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
Studies in Fuzziness and Soft Computing, 2002
In this chapter we focus on the Wisconsin breast cancer diagnosis (WBCD) problem, combining two m... more In this chapter we focus on the Wisconsin breast cancer diagnosis (WBCD) problem, combining two methodologies-fuzzy systems and evolutionary algorithms-to automatically produce diagnostic systems. We present two hybrid approaches: (1) a fuzzy-genetic algorithm, and (2) Fuzzy CoCo, a novel cooperative coevolutionary approach to fuzzy modeling. Both methods produce systems exhibiting high classification performance, and which are also human-interpretable. Fuzzy CoCo obtains higher-performance systems than the standard fuzzy-genetic approach while expending less computational effort.
FUZZ-IEEE'99. 1999 IEEE International Fuzzy Systems. Conference Proceedings (Cat. No.99CH36315), 1999
The automatic diagnosis of breast cancer is an important, real-world medical problem. In this pap... more The automatic diagnosis of breast cancer is an important, real-world medical problem. In this paper we focus on the Wisconsin breast cancer diagnosis (WBCD) problem, combining two methodo-logies|fuzzy systems and evolutionary algorithms|so as to automatically produce diagnostic systems. We nd that our fuzzy-genetic approach produces systems exhibiting the highest classi cation performance shown to date, and which are also (human-)interpretable.
Proceedins of the 14th ACM Great Lakes symposium on VLSI - GLSVLSI '04, 2004
This paper presents an architectural proposal for a hardware-based interval type-2 fuzzy inferenc... more This paper presents an architectural proposal for a hardware-based interval type-2 fuzzy inference system. First, it presents a computational model which considers parallel inference processing and type reduction based on computing inner and outer bound sets. Taking into account this model, we conceived a hardware architecture with several pipeline stages for full parallel execution of type-2 fuzzy inferences. The architectural proposal is used for specifying a type-2 fuzzy processor with reconfigurable rule base, which is implemented over FPGA technology. Implementation results show that this processor performs more than 30 millions of type-2 fuzzy inferences per second. Categories and Subject Descriptors B.7.1 [Integrated Circuits]: Types and design styles-algorithms implemented in hardware, gate arrays.
Pro-Two: a hardware based platform for real time type-2 fuzzy inference
2004 IEEE International Conference on Fuzzy Systems (IEEE Cat. No.04CH37542), 2004
Page 1. 25-29 July, 2004 * Budapest, Hungary Pro-Two: A Hardware Based Platform For Real Time Typ... more Page 1. 25-29 July, 2004 * Budapest, Hungary Pro-Two: A Hardware Based Platform For Real Time Type-2 Fuzzy Inference Miguel A. Melgarejo Antonio Garcia R. Carlos A. Pefia-Reyes Department of Department of Logic Systems Laboratory ...
Memorias del Congreso Internacional de Inteligencia Computacional, Montería, Colombia, Agosto 10-12 de 2005
Congreso Internacional de Inteligencia Computacional, 2005
Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512), 2000
Coevolutionary algorithms have received increased attention in the past few years within the doma... more Coevolutionary algorithms have received increased attention in the past few years within the domain of evolutionary computation. In this paper, we combine the search power of coevolutionary computation with the expressive power of fuzzy systems, introducing a novel algorithm, Fuzzy CoCo: Fuzzy Cooperative Coevolution. We demonstrate the efficacy of Fuzzy CoCo by applying it to a hard, real-world problem-breast cancer diagnosisobtaining the best results to date while expending less computational effort than formerly.
Evolving very-compact fuzzy models for gene expression data analysis
2012 IEEE 12th International Conference on Bioinformatics & Bioengineering (BIBE), 2012
ABSTRACT Selecting predicitve gene pools from thousands of gene expression values is one of the m... more ABSTRACT Selecting predicitve gene pools from thousands of gene expression values is one of the main tasks in microarray data analysis. For this purpose multivariate techniques have proven much better, in terms of predicitve value and biological relevance, than univariate techniques as they are able to capture relevant relationships and interactions between genes. An additional goal for gene-expression profiling is finding models that, besides being predictive, are also understandable so as they can provide some insight on the underlying mechanisms. Models based on fuzzy logic might, potentially, exhibit both characteristics. However, accuracy and interpretability are usually contradictory objectives, and one must accept a trade off between them. Indeed, literature shows that the approaches based on fuzzy logic may be divided in two groups: accurate but complex models (i.e, with many rules using many variables per rule) on one hand, and models with only few short rules (thus, interpretable) but exhibiting limited accuracy. We present in this paper the application of Fuzzy CoCo, our cooperative coevolutionary fuzzy modelling approach, in order to deal efficiently with the accuracy-interpretability tradeoff. Fuzzy CoCo is able to find very compact fuzzy models, in terms of number of rules and number of variables per rule, while still exhibiting high predictive power. To validate the performance of our approach, we tested Fuzzy CoCo on four known data sets addressing each one a form of cancer: Leukemia, colon, lung, and prostate. We compared our results-in terms of maximum number of rules, number of variables per rule, and accuracy-with those of other similar works (i.e., based on fuzzy logic). Our models reached similar or better accuracy while being considerably smaller.
Studies in Fuzziness and Soft Computing, 2003
In this chapter we present Fuzzy CoCo, a fuzzy modeling technique based on cooperative coevolutio... more In this chapter we present Fuzzy CoCo, a fuzzy modeling technique based on cooperative coevolution, conceived to provide high numeric precision (accuracy) while incurring as little a loss of linguistic descriptive power (interpretability) as possible. The search for interpretability is represented by several constraints taken into account when designing the evolutionary algorithm, which induce the drive for accuracy. Interpretability-oriented fuzzy modeling must conduct two separate but intertwined search processes: (1) the search for membership functions, and (2) the search for rules. Towards this end, Fuzzy CoCo employs two coevolving species: database (membership functions) and rule base. Coevolution allows to overcome limitations presented by single-population evolutionary algorithms when confronted with fuzzy modeling, including stagnation, convergence to local optima, and computational costliness. We demonstrate the efficacy of Fuzzy CoCo by applying it to a hard, real-world problem-prediction of breast-cancer malignancyobtaining excellent results.
Competitive intelligence and patent analysis in drug discovery
Drug discovery today. Technologies, 2005
Patents are a major source of information in drug discovery and, when properly processed and anal... more Patents are a major source of information in drug discovery and, when properly processed and analyzed, can yield a wealth of information on competitors activities, R&D trends, emerging fields, collaborations, among others. This review discusses the current state-of-the-art in textual data analysis and exploration methods as applied to patent analysis.:
Lecture Notes in Computer Science, 2004
In this paper we present a functional model of a spiking neuron intended for hardware implementat... more In this paper we present a functional model of a spiking neuron intended for hardware implementation. Some features of biological spiking neurons are abstracted, while preserving the functionality of the network, in order to define an architecture with low implementation cost in field programmable gate arrays (FPGAs). Adaptation of synaptic weights is implemented with hebbian learning. As an example application we present a frequency discriminator to verify the computing capabilities of a generic network of our neuron model.
Lecture Notes in Computer Science, 2005
In this contribution, we describe a hardware platform for evolving a fuzzy system by using Fuzzy ... more In this contribution, we describe a hardware platform for evolving a fuzzy system by using Fuzzy CoCoa cooperative coevolutionary methodology for fuzzy system design-in order to speed up both evolution and execution. Reconfigurable hardware arises between hardware and software solutions providing a trade-off between flexibility and performance. We present an architecture that exploits the dynamic partial reconfiguration capabilities of recent FPGAs so as to provide adaptation at two different levels: major structural changes and fuzzy parameter tuning.
Stability of Feature Selection Methods: A Study of Metrics Across Different Gene Expression Datasets
Bioinformatics and Biomedical Engineering, 2020
Analysis of gene-expression data often requires that a gene (feature) subset is selected and many... more Analysis of gene-expression data often requires that a gene (feature) subset is selected and many feature selection (FS) methods have been devised. However, FS methods often generate different lists of features for the same dataset and users then have to choose which list to use. One approach to support this choice is to apply stability metrics on the generated lists and selecting lists on that base. The aim of this study is to investigate the behavior of stability metrics applied to feature subsets generated by FS methods. The experiments in this work explore a plethora of gene expression datasets, FS methods, and expected number of features to compare several stability metrics. The stability metrics have been used to compare five feature selection methods (SVM, SAM, ReliefF, RFE + RF and LIMMA) on gene expression datasets from the EBI repository. Results show that the studied stability metrics display a high amount of variability. The reason behind this is not clear yet and is being further investigated. The final objective of the research, that is to define how to select a FS method, is an ongoing work whose partial findings are reported herein.
Bioinformatics and Biomedical Engineering, 2017
The emergence and rapid dissemination of antibiotic resistance, worldwide, threatens medical prog... more The emergence and rapid dissemination of antibiotic resistance, worldwide, threatens medical progress and calls for innovative approaches for the management of multidrug resistant infections. Phage-therapy, i.e., the use of viruses (phages) that specifically infect and kill bacteria during their life cycle, is a re-emerging and promising alternative to solve this problem. The success of phage therapy mainly relies on the exact matching between the target pathogenic bacteria and the therapeutic phage. Currently, there are only a few tools or methodologies that efficiently predict phage-bacteria interactions suitable for the phage therapy, and the pairs phage-bacterium are thus empirically tested in laboratory. In this paper we present an original methodology, based on an ensemble-learning approach, to predict whether or not a given pair of phage-bacteria would interact. Using publicly available information from Genbank and phagesdb.org, we assembled a dataset containing more than two thousand phage-bacterium interactions with their corresponding genomes. A set of informative features, extracted from these genomes, form the base of the quantitative datasets used to train our predictive models. These features include the distribution of predicted protein-protein interaction scores, as well as the amino acid frequency, the chemical composition, and the molecular weight of such proteins. Using an independent test dataset to evaluate the performance of our methodology, our approach gets encouraging performance with more than 90% of accuracy, specificity, and sensitivity.
Agent-based modeling of mesenchymal stem cells on a 3D-printed bio-device for the regenerative treatment of the infarcted myocardium
A myocardial infarction (MI) implies an obstruction of the blood flow due to a blockade in the co... more A myocardial infarction (MI) implies an obstruction of the blood flow due to a blockade in the coronary artery, which results in the cellular necrosis of the myocardium due to ischemia. Mesenchymal stem cells (MSCs) are non-differentiated cells with the ability to differentiate into any cell type. A regenerative treatment has been proposed for the regeneration of the infarcted myocardium, as the existing ones do not aim at the regeneration of the necrosed tissue. The treatment proposed is based on the design and construction of a 3D-printed biodevice, containing MSCs and other cell types, which can be later implanted on the patient's myocardium.A good understanding of the behavior of the cells in the biodevice is important to guide the treatment research. Thereby, the development of a computational model of the behavior of the MSC and other cell types used in the 3D-printed biodevice is carried out, allowing to better understand its dynamics and the effect that different variabl...
This paper describes a genetic-fuzzy approach for controlling a nonlinear model of the HIV immuno... more This paper describes a genetic-fuzzy approach for controlling a nonlinear model of the HIV immunology. The approach is set up to find mamdani fuzzy controllers capable of boosting the immune response while reducing the systemic cost to the body due to the use of high efficacy drugs. General aspects of the genetic fuzzy system are described while special emphasis is given to control results. The best controller found from several evolutionary runs is analyzed in terms of control performance and it is compared with a traditional medical procedure. Simulation results show that the fuzzy control approach provides a better long-term treatment.
Data, 2019
Regenerative medicine involves methods to control and modify normal tissue repair processes. Poly... more Regenerative medicine involves methods to control and modify normal tissue repair processes. Polymer and cell constructs are under research to create tissue that replaces the affected area in cardiac tissue after myocardial infarction (MI). The aim of the present study is to evaluate the behavior of differentiated and undifferentiated mesenchymal stem cells (MSCs) in vitro and in silico and to compare the results that both offer when it comes to the design process of biodevices for the treatment of infarcted myocardium in biomodels. To assess in vitro behavior, MSCs are isolated from rat bone marrow and seeded undifferentiated and differentiated in multiple scaffolds of a gelled biomaterial. Subsequently, cell behavior is evaluated by trypan blue and fluorescence microscopy, which showed that the cells presented high viability and low cell migration in the biomaterial. An agent-based model intended to reproduce as closely as possible the behavior of individual MSCs by simulating cel...
BMC Bioinformatics, 2018
Background: Antibiotic resistance and its rapid dissemination around the world threaten the effic... more Background: Antibiotic resistance and its rapid dissemination around the world threaten the efficacy of currently-used medical treatments and call for novel, innovative approaches to manage multi-drug resistant infections. Phage therapy, i.e., the use of viruses (phages) to specifically infect and kill bacteria during their life cycle, is one of the most promising alternatives to antibiotics. It is based on the correct matching between a target pathogenic bacteria and the therapeutic phage. Nevertheless, correctly matching them is a major challenge. Currently, there is no systematic method to efficiently predict whether phage-bacterium interactions exist and these pairs must be empirically tested in laboratory. Herein, we present our approach for developing a computational model able to predict whether a given phage-bacterium pair can interact based on their genome. Results: Based on public data from GenBank and phagesDB.org, we collected more than a thousand positive phage-bacterium interactions with their complete genomes. In addition, we generated putative negative (i.e., non-interacting) pairs. We extracted, from the collected genomes, a set of informative features based on the distribution of predictive protein-protein interactions and on their primary structure (e.g. amino-acid frequency, molecular weight and chemical composition of each protein). With these features, we generated multiple candidate datasets to train our algorithms. On this base, we built predictive models exhibiting predictive performance of around 90% in terms of F1-score, sensitivity, specificity, and accuracy, obtained on the test set with 10-fold cross-validation. Conclusion: These promising results reinforce the hypothesis that machine learning techniques may produce highly-predictive models accelerating the search of interacting phage-bacteria pairs.
IEEE Computational Intelligence Magazine, 2007
2006 IEEE International Conference on Fuzzy Systems, 2006
This paper presents a genetic fuzzy system approach to control a nonlinear dynamic model of the H... more This paper presents a genetic fuzzy system approach to control a nonlinear dynamic model of the HIV infection. The system is conceived to find mamdani fuzzy controllers that are capable of boosting the immune response while reducing the impact on the body because of the use of potentially toxic medicaments. General aspects of the used approach are described while special emphasis is given to the evolutionary mechanism. The best solution found from several evolutionary runs is analyzed in terms of control and interpretability issues.
Exploration of multiclass and one-class learning methods for prediction of phage-bacteria interaction at strain level
2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
Studies in Fuzziness and Soft Computing, 2002
In this chapter we focus on the Wisconsin breast cancer diagnosis (WBCD) problem, combining two m... more In this chapter we focus on the Wisconsin breast cancer diagnosis (WBCD) problem, combining two methodologies-fuzzy systems and evolutionary algorithms-to automatically produce diagnostic systems. We present two hybrid approaches: (1) a fuzzy-genetic algorithm, and (2) Fuzzy CoCo, a novel cooperative coevolutionary approach to fuzzy modeling. Both methods produce systems exhibiting high classification performance, and which are also human-interpretable. Fuzzy CoCo obtains higher-performance systems than the standard fuzzy-genetic approach while expending less computational effort.
FUZZ-IEEE'99. 1999 IEEE International Fuzzy Systems. Conference Proceedings (Cat. No.99CH36315), 1999
The automatic diagnosis of breast cancer is an important, real-world medical problem. In this pap... more The automatic diagnosis of breast cancer is an important, real-world medical problem. In this paper we focus on the Wisconsin breast cancer diagnosis (WBCD) problem, combining two methodo-logies|fuzzy systems and evolutionary algorithms|so as to automatically produce diagnostic systems. We nd that our fuzzy-genetic approach produces systems exhibiting the highest classi cation performance shown to date, and which are also (human-)interpretable.
Proceedins of the 14th ACM Great Lakes symposium on VLSI - GLSVLSI '04, 2004
This paper presents an architectural proposal for a hardware-based interval type-2 fuzzy inferenc... more This paper presents an architectural proposal for a hardware-based interval type-2 fuzzy inference system. First, it presents a computational model which considers parallel inference processing and type reduction based on computing inner and outer bound sets. Taking into account this model, we conceived a hardware architecture with several pipeline stages for full parallel execution of type-2 fuzzy inferences. The architectural proposal is used for specifying a type-2 fuzzy processor with reconfigurable rule base, which is implemented over FPGA technology. Implementation results show that this processor performs more than 30 millions of type-2 fuzzy inferences per second. Categories and Subject Descriptors B.7.1 [Integrated Circuits]: Types and design styles-algorithms implemented in hardware, gate arrays.
Pro-Two: a hardware based platform for real time type-2 fuzzy inference
2004 IEEE International Conference on Fuzzy Systems (IEEE Cat. No.04CH37542), 2004
Page 1. 25-29 July, 2004 * Budapest, Hungary Pro-Two: A Hardware Based Platform For Real Time Typ... more Page 1. 25-29 July, 2004 * Budapest, Hungary Pro-Two: A Hardware Based Platform For Real Time Type-2 Fuzzy Inference Miguel A. Melgarejo Antonio Garcia R. Carlos A. Pefia-Reyes Department of Department of Logic Systems Laboratory ...
Memorias del Congreso Internacional de Inteligencia Computacional, Montería, Colombia, Agosto 10-12 de 2005
Congreso Internacional de Inteligencia Computacional, 2005
Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512), 2000
Coevolutionary algorithms have received increased attention in the past few years within the doma... more Coevolutionary algorithms have received increased attention in the past few years within the domain of evolutionary computation. In this paper, we combine the search power of coevolutionary computation with the expressive power of fuzzy systems, introducing a novel algorithm, Fuzzy CoCo: Fuzzy Cooperative Coevolution. We demonstrate the efficacy of Fuzzy CoCo by applying it to a hard, real-world problem-breast cancer diagnosisobtaining the best results to date while expending less computational effort than formerly.
Evolving very-compact fuzzy models for gene expression data analysis
2012 IEEE 12th International Conference on Bioinformatics & Bioengineering (BIBE), 2012
ABSTRACT Selecting predicitve gene pools from thousands of gene expression values is one of the m... more ABSTRACT Selecting predicitve gene pools from thousands of gene expression values is one of the main tasks in microarray data analysis. For this purpose multivariate techniques have proven much better, in terms of predicitve value and biological relevance, than univariate techniques as they are able to capture relevant relationships and interactions between genes. An additional goal for gene-expression profiling is finding models that, besides being predictive, are also understandable so as they can provide some insight on the underlying mechanisms. Models based on fuzzy logic might, potentially, exhibit both characteristics. However, accuracy and interpretability are usually contradictory objectives, and one must accept a trade off between them. Indeed, literature shows that the approaches based on fuzzy logic may be divided in two groups: accurate but complex models (i.e, with many rules using many variables per rule) on one hand, and models with only few short rules (thus, interpretable) but exhibiting limited accuracy. We present in this paper the application of Fuzzy CoCo, our cooperative coevolutionary fuzzy modelling approach, in order to deal efficiently with the accuracy-interpretability tradeoff. Fuzzy CoCo is able to find very compact fuzzy models, in terms of number of rules and number of variables per rule, while still exhibiting high predictive power. To validate the performance of our approach, we tested Fuzzy CoCo on four known data sets addressing each one a form of cancer: Leukemia, colon, lung, and prostate. We compared our results-in terms of maximum number of rules, number of variables per rule, and accuracy-with those of other similar works (i.e., based on fuzzy logic). Our models reached similar or better accuracy while being considerably smaller.
Studies in Fuzziness and Soft Computing, 2003
In this chapter we present Fuzzy CoCo, a fuzzy modeling technique based on cooperative coevolutio... more In this chapter we present Fuzzy CoCo, a fuzzy modeling technique based on cooperative coevolution, conceived to provide high numeric precision (accuracy) while incurring as little a loss of linguistic descriptive power (interpretability) as possible. The search for interpretability is represented by several constraints taken into account when designing the evolutionary algorithm, which induce the drive for accuracy. Interpretability-oriented fuzzy modeling must conduct two separate but intertwined search processes: (1) the search for membership functions, and (2) the search for rules. Towards this end, Fuzzy CoCo employs two coevolving species: database (membership functions) and rule base. Coevolution allows to overcome limitations presented by single-population evolutionary algorithms when confronted with fuzzy modeling, including stagnation, convergence to local optima, and computational costliness. We demonstrate the efficacy of Fuzzy CoCo by applying it to a hard, real-world problem-prediction of breast-cancer malignancyobtaining excellent results.
Competitive intelligence and patent analysis in drug discovery
Drug discovery today. Technologies, 2005
Patents are a major source of information in drug discovery and, when properly processed and anal... more Patents are a major source of information in drug discovery and, when properly processed and analyzed, can yield a wealth of information on competitors activities, R&D trends, emerging fields, collaborations, among others. This review discusses the current state-of-the-art in textual data analysis and exploration methods as applied to patent analysis.:
Lecture Notes in Computer Science, 2004
In this paper we present a functional model of a spiking neuron intended for hardware implementat... more In this paper we present a functional model of a spiking neuron intended for hardware implementation. Some features of biological spiking neurons are abstracted, while preserving the functionality of the network, in order to define an architecture with low implementation cost in field programmable gate arrays (FPGAs). Adaptation of synaptic weights is implemented with hebbian learning. As an example application we present a frequency discriminator to verify the computing capabilities of a generic network of our neuron model.
Lecture Notes in Computer Science, 2005
In this contribution, we describe a hardware platform for evolving a fuzzy system by using Fuzzy ... more In this contribution, we describe a hardware platform for evolving a fuzzy system by using Fuzzy CoCoa cooperative coevolutionary methodology for fuzzy system design-in order to speed up both evolution and execution. Reconfigurable hardware arises between hardware and software solutions providing a trade-off between flexibility and performance. We present an architecture that exploits the dynamic partial reconfiguration capabilities of recent FPGAs so as to provide adaptation at two different levels: major structural changes and fuzzy parameter tuning.