Comparing early and late data fusion methods for gene expression prediction

Abstract  The most basic molecular mechanism enabling a living cell to dynamically adapt to variation occurring in its intra and extracellular
environment is constituted by its ability to regulate the expression of many of its genes. At biom…

Abstract  

The most basic molecular mechanism enabling a living cell to dynamically adapt to variation occurring in its intra and extracellular
environment is constituted by its ability to regulate the expression of many of its genes. At biomolecular level, this ability
is mainly due to interactions occurring between regulatory motifs located in the core promoter regions and the transcription
factors. A crucial question investigated by recently published works is if, and at what extent, the transcription patterns
of large sets of genes can be predicted using only information encoded in the promoter regions. Even if encouraging results
were obtained in gene expression patterns prediction experiments the assumption that all the signals required for the regulation
of gene expression are contained in the gene promoter regions is an oversimplification as pointed out by recent findings demonstrating
the existence of many regulatory levels involved in the fine modulation of gene transcription levels. In this contribution,
we investigate the potential improvement in gene expression prediction performances achievable by using early and late data
integration methods in order to provide a complete overview of the capabilities of data fusion approaches in a problem that
can be annoverated among the most difficult in modern bioinformatics.

  • Content Type Journal Article
  • Pages 1-8
  • DOI 10.1007/s00500-010-0599-6
  • Authors
    • Matteo Re, Universitá degli studi di Milano Dipartimento di Scienze dell’Informazione, DSI via Comelico 39 Milan Italy

Biocircuit design through engineering bacterial logic gates

Abstract  Designing synthetic biocircuits to perform desired purposes is a scientific field that has exponentially grown over the past
decade. The advances in genome sequencing, bacteria gene regulatory networks, as well as the further knowl…

Abstract  

Designing synthetic biocircuits to perform desired purposes is a scientific field that has exponentially grown over the past
decade. The advances in genome sequencing, bacteria gene regulatory networks, as well as the further knowledge of intraspecies
bacterial communication through quorum sensing signals are the starting point for this work. Although biocircuits are mostly
developed in a single cell, here we propose a model in which every bacterium is considered to be a single logic gate and chemical
cell-to-cell connections are engineered to control circuit function. Having one genetically modified bacterial strain per
logic process would allow us to develop circuits with different behaviors by mixing the populations instead of re-programming
the whole genetic network within a single strain. Two principal advantages of this procedure are highlighted. First, the fully
connected circuits obtained where every cellgate is able to communicate with all the rest. Second, the resistance to the noise
produced by inappropriate gene expression. This last goal is achieved by modeling thresholds for input signals. Thus, if the
concentration of input does not exceed the threshold, it is ignored by the logic function of the gate.

  • Content Type Journal Article
  • Pages 1-9
  • DOI 10.1007/s11047-010-9184-2
  • Authors
    • Angel Goñi-Moreno, Universidad Politécnica de Madrid Grupo de Computación Natural, Facultad de Informática 28660 Madrid Spain
    • Miguel Redondo-Nieto, Universidad Autónoma de Madrid Depto. Biología, Facultad de Ciencias 28049 Madrid Spain
    • Fernando Arroyo, Universidad Politécnica de Madrid Depto. de Lenguajes, Proyectos y Sistemas Informáticos, Escuela Universitaria de Informática 28031 Madrid Spain
    • Juan Castellanos, Universidad Politécnica de Madrid Artificial Intelligence Department, Facultad de Informática Campus de Montegancedo, Boadilla del Monte s/n 28660 Madrid Spain

The quantification of pollutants in drinking water by use of artificial neural networks

Abstract  Drinking water attained from aquifers (ground water) is susceptible to contamination from a wide variety of sources. The importance
of ensuring that the water is of high quality is paramount. Multivariate calibration in conjunction…

Abstract  

Drinking water attained from aquifers (ground water) is susceptible to contamination from a wide variety of sources. The importance
of ensuring that the water is of high quality is paramount. Multivariate calibration in conjunction with analytical techniques
can assist in qualifying and quantifying a wide range of pollutants. These can be divided into two types: inorganic and organic.
The former typically includes heavy metals such as cadmium and lead; the latter includes a range of compounds such as pesticides
and by-products of industrial processes such as oil refining. This article presents the application of the well known nature-inspired
paradigm of artificial neural networks (ANNs) for the quantitative determination of inorganic pollutants (namely cadmium,
lead and copper) and organic pollutants (namely anthracene, phenanthrene and naphthalene) from multivariate analytical data
acquired from the samples. The success of the determination of the pollutants via ANNs is reported in terms of the overall
root mean square error of prediction (RMSEP) which is an accepted measure of the difference between the predicted concentrations
and the actual concentrations. The work represents a good example of nature-inspired methods being used to solve a genuine
environmental problem.

  • Content Type Journal Article
  • Pages 1-14
  • DOI 10.1007/s11047-010-9185-1
  • Authors
    • Michael Cauchi, Cranfield University Cranfield Health Bedfordshire MK43 0AL UK
    • Luca Bianco, Cranfield University Cranfield Health Bedfordshire MK43 0AL UK
    • Conrad Bessant, Cranfield University Cranfield Health Bedfordshire MK43 0AL UK

Advances in Computational Intelligence and Bioinformatics

Advances in Computational Intelligence and Bioinformatics
Content Type Journal ArticlePages 1-2DOI 10.1007/s00500-010-0595-xAuthors
Francesco Masulli, Università di Genova DISI, Dipartimento di Informatica e Scienze, dell’Informazione Via Dodecan…

Advances in Computational Intelligence and Bioinformatics

  • Content Type Journal Article
  • Pages 1-2
  • DOI 10.1007/s00500-010-0595-x
  • Authors
    • Francesco Masulli, Università di Genova DISI, Dipartimento di Informatica e Scienze, dell’Informazione Via Dodecaneso 35 Genoa Italy
    • Roberto Tagliaferri, Università di Salerno NeuRoNe Lab, DMI, Dipartimento di Matematica e Informatica Via Ponte don Melillo Fisciano (SA) Italy

History mechanism supported differential evolution for chess evaluation function tuning

Abstract  This paper presents a differential evolution (DE) based approach to chess evaluation function tuning. DE with opposition-based
optimization is employed and upgraded with a history mechanism to improve the evaluation of individuals …

Abstract  

This paper presents a differential evolution (DE) based approach to chess evaluation function tuning. DE with opposition-based
optimization is employed and upgraded with a history mechanism to improve the evaluation of individuals and the tuning process.
The general idea is based on individual evaluations according to played games through several generations and different environments.
We introduce a new history mechanism which uses an auxiliary population containing good individuals. This new mechanism ensures
that good individuals remain within the evolutionary process, even though they died several generations back and later can
be brought back into the evolutionary process. In such a manner the evaluation of individuals is improved and consequently
the whole tuning process.

  • Content Type Journal Article
  • Pages 1-17
  • DOI 10.1007/s00500-010-0593-z
  • Authors
    • B. Bošković, Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova ul. 17, 2000 Maribor, Slovenia
    • J. Brest, Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova ul. 17, 2000 Maribor, Slovenia
    • A. Zamuda, Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova ul. 17, 2000 Maribor, Slovenia
    • S. Greiner, Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova ul. 17, 2000 Maribor, Slovenia
    • V. Žumer, Faculty of Electrical Engineering and Computer Science, University of Maribor, Smetanova ul. 17, 2000 Maribor, Slovenia

Rule acquisition and attribute reduction in real decision formal contexts

Abstract  Formal Concept Analysis of real set formal contexts is a generalization of classical formal contexts. By dividing the attributes
into condition attributes and decision attributes, the notion of real decision formal contexts is intr…

Abstract  

Formal Concept Analysis of real set formal contexts is a generalization of classical formal contexts. By dividing the attributes
into condition attributes and decision attributes, the notion of real decision formal contexts is introduced. Based on an
implication mapping, problems of rule acquisition and attribute reduction of real decision formal contexts are examined. The
extraction of “if–then” rules from the real decision formal contexts, and the approach to attribute reduction of the real
decision formal contexts are discussed. By the proposed approach, attributes which are non-essential to the maximal s rules or l rules (to be defined later in the text) can be removed. Furthermore, discernibility matrices and discernibility functions
for computing the attribute reducts of the real decision formal contexts are constructed to determine all attribute reducts
of the real set formal contexts without affecting the results of the acquired maximal s rules or l rules.

  • Content Type Journal Article
  • Pages 1-14
  • DOI 10.1007/s00500-010-0578-y
  • Authors
    • Hong-Zhi Yang, Xi’an Jiaotong University Faculty of Science Xi’an 710049 Shaan’xi People’s Republic of China
    • Leung Yee, University of Hong Kong Department of Geography and Resource Management, Center for Environmental Policy and Resource Management Hong Kong People’s Republic of China
    • Ming-Wen Shao, Shihezi University College of Information Science and Technology Shihezi 832000 Xinjiang People’s Republic of China

Covariance matrix self-adaptation evolution strategies and other metaheuristic techniques for neural adaptive learning

Abstract  A covariance matrix self-adaptation evolution strategy (CMSA-ES) was compared with several metaheuristic techniques for multilayer
perceptron (MLP)-based function approximation and classification. Function approximation was based o…

Abstract  

A covariance matrix self-adaptation evolution strategy (CMSA-ES) was compared with several metaheuristic techniques for multilayer
perceptron (MLP)-based function approximation and classification. Function approximation was based on simulations of several
2D functions and classification analysis was based on nine cancer DNA microarray data sets. Connection weight learning by
MLPs was carried out using genetic algorithms (GA–MLP), covariance matrix self-adaptation-evolution strategies (CMSA-ES–MLP),
back-propagation gradient-based learning (MLP), particle swarm optimization (PSO–MLP), and ant colony optimization (ACO–MLP).
During function approximation runs, input-side activation functions evaluated included linear, logistic, tanh, Hermite, Laguerre,
exponential, and radial basis functions, while the output-side function was always linear. For classification, the input-side
activation function was always logistic, while the output-side function was always regularized softmax. Self-organizing maps
and unsupervised neural gas were used to reduce dimensions of original gene expression input features used in classification.
Results indicate that for function approximation, use of Hermite polynomials for activation functions at hidden nodes with
CMSA-ES–MLP connection weight learning resulted in the greatest fitness levels. On average, the most elite chromosomes were
observed for MLP (

MSE=0.4977

), CMSA-ES–MLP (0.6484), PSO–MLP (0.7472), ACO–MLP (1.3471), and GA–MLP (1.4845). For classification analysis, overall average
performance of classifiers used was 92.64% (CMSA-ES–MLP), 92.22% (PSO–MLP), 91.30% (ACO–MLP), 89.36% (MLP), and 60.72% (GA–MLP).
We have shown that a reliable approach to function approximation can be achieved through application of MLP connection weight
learning when the assumed function is unknown. In this scenario, the MLP architecture itself defines the equation used for
solving the unknown parameters relating input and output target values. A major drawback of implementing CMSA-ES into an MLP
is that when the number of MLP weights is large, the

O(N3)

Cholesky factorization becomes a bottleneck for performance. As an alternative, feature reduction using SOM and NG can greatly
enhance performance of CMSA-ES–MLP by reducing

N.

Future research into the speeding up of Cholesky factorization for CMSA-ES will be helpful in overcoming time complexity
problems related to a large number of connection weights.

  • Content Type Journal Article
  • Pages 1-13
  • DOI 10.1007/s00500-010-0598-7
  • Authors
    • Leif E. Peterson, Center for Biostatistics, The Methodist Hospital Research Institute, Houston, TX 77030, USA

The Perils and Pleasures of Interdisciplinarity

Yesterday David E. Goldberg gave a talk on The Perils and Pleasures of Interdisciplinarity at a Workshop on the Challenges in Top-Down, Bottom-up and Computational Approaches in Synthetic Biology. The slides of the talk are available via slideshare:
The Perils & Pleasures of Interdisciplinarity
View more presentations from deg511.

For a related post, visit IlliGAL Blogging.

Yesterday David E. Goldberg gave a talk on The Perils and Pleasures of Interdisciplinarity at a Workshop on the Challenges in Top-Down, Bottom-up and Computational Approaches in Synthetic Biology. The slides of the talk are available via slideshare:

The Perils & Pleasures of Interdisciplinarity

View more presentations from deg511.

For a related post, visit IlliGAL Blogging.

Hedges and successors in basic algebras

Abstract  The concept of hedge was introduced by Zadeh in the sake to amplify true values of linguistic terms. It was used by Bělohlávek
and Vychodil for formal concept analysis of unsharp reasoning. The concept of successor was introduced…

Abstract  

The concept of hedge was introduced by Zadeh in the sake to amplify true values of linguistic terms. It was used by Bělohlávek
and Vychodil for formal concept analysis of unsharp reasoning. The concept of successor was introduced by Caicedo and Cignoli
for study of intuitionistic connectives and used by San Martín, Castiglioni, Menni and Sagastume in Heyting algebras. Since
basic algebras form an algebraic tool for simultaneous treaty of many-valued logics and logics of quantum mechanics, it arises
a natural question of generalization of these concepts also for basic algebras. This motivated our investigations on hedges
and successors.

  • Content Type Journal Article
  • Pages 1-6
  • DOI 10.1007/s00500-010-0570-6
  • Authors
    • Ivan Chajda, Palacký University Olomouc Department of Algebra and Geometry, Faculty of Sciences třída 17. listopadu 12 771 46 Olomouc Czech Republic

The perils and pleasures of interdisciplinarity

IlliGAL director David E. Goldberg just gave a talk on “The Perils and Pleasures of Interdisciplinarity” at a Workshop on the Challenges in Top-Down, Bottom-up and Computational Approaches in Synthetic Biology.  The talk is available in the viewer below:Related talks … Continue reading →

IlliGAL director David E. Goldberg just gave a talk on “The Perils and Pleasures of Interdisciplinarity” at a Workshop on the Challenges in Top-Down, Bottom-up and Computational Approaches in Synthetic Biology.  The talk is available in the viewer below:

Related talks are available here.