Biomedical subjects

Yves Moreau

Publications and source records attributed to Yves Moreau.

29 records · Page 2Linked to original sources

PubMed2003

Toucan: deciphering the cis-regulatory logic of coregulated genes.

TOUCAN is a Java application for the rapid discovery of significant cis-regulatory elements from sets of coexpressed or coregulated genes. Biologists can automatically (i) retrieve genes and intergenic regions, (ii) identify putative regulatory regions, (iii) score sequences for known transcription factor binding sites, (iv) identify candidate motifs for unknown binding sites, and (v) detect those statistically over-represented sites that are characteristic for a gene set. Genes or intergenic regions are retrieved from Ensembl or EMBL, together with orthologs and supporting information. Orthologs are aligned and syntenic regions are selected as candidate regulatory regions. Putative sites for known transcription factors are detected using our MotifScanner, which scores position weight matrices using a probabilistic model. New motifs are detected using our MotifSampler based on Gibbs sampling. Binding sites characteristic for a gene set--and thus statistically over-represented with respect to a reference sequence set--are found using a binomial test. We have validated Toucan by analyzing muscle-specific genes, liver-specific genes and E2F target genes; we have easily detected many known binding sites within intergenic DNA and identified new biologically plausible sites for known and unknown transcription factors. Software available at http://www.esat.kuleuven.ac. be/ approximately dna/BioI/Software.html.

Algorithms↗

PubMed2003

Comparison and meta-analysis of microarray data: from the bench to the computer desk.

The upcoming availability of public microarray repositories and of large compendia of gene expression information opens up a new realm of possibilities for microarray data analysis. An essential challenge is the efficient integration of microarray data generated by different research groups on different array platforms. This review focuses on the problems associated with this integration, which are: (1) the efficient access to and exchange of microarray data; (2) the validation and comparison of data from different platforms (cDNA and short and long oligonucleotides); and (3) the integrated statistical analysis of multiple data sets.

Databases, Genetic↗

PubMed2003

Gene profiling of hippocampal neuronal culture.

We performed mRNA expression profiling of mouse primary hippocampal neurones undergoing differentiation in vitro. We show that 2314 genes significantly changed expression during neuronal differentiation. The temporal resolution of our experiment (six time points) permits us to distinguish between gene expression patterns characteristic for the axonal and for the dendritic stages of neurite outgrowth. Cluster analysis reveals that, in the process of in vitro neuronal differentiation, a high level of expression of genes involved in the synthesis of DNA and proteins precedes the up regulation of genes involved in protein transport, energy generation and synaptic functions. We report in detail changes in gene expression for genes involved in the synaptic vesicle cycle. Data for other genes can be accessed at our website. We directly compare expression of 475 genes in the differentiating neurones and the developing mouse hippocampus. We demonstrate that the program of gene expression is accelerated in vitro as compared to the situation in vivo. When this factor is accounted for, the gene expression profiles in vitro and in vivo become very similar (median gene-wise correlation 0.787). Apparently once the cells have taken a neuronal fate, the further program of gene expression is largely independent of histological or anatomical context. Our results also demonstrate that a comparison across the two experimental platforms (cDNA microarrays and oligonucleotide chips) and across different biological paradigms is feasible.

Animals↗

PubMed2003

Computational detection of cis -regulatory modules.

MOTIVATION: The transcriptional regulation of a metazoan gene depends on the cooperative action of multiple transcription factors that bind to cis-regulatory modules (CRMs) located in the neighborhood of the gene. By integrating multiple signals, CRMs confer an organism specific spatial and temporal rate of transcription. RESULTS: Based on the hypothesis that genes that are needed in exactly the same conditions might share similar regulatory switches, we have developed a novel methodology to find CRMs in a set of coexpressed or coregulated genes. The ModuleSearcher algorithm finds for a given gene set the best scoring combination of transcription factor binding sites within a sequence window using an A(*)procedure for tree searching. To keep the level of noise low, we use DNA sequences that are most likely to contain functional cis-regulatory information, namely conserved regions between human and mouse orthologous genes. The ModuleScanner performs genomic searches with a predicted CRM or with a user-defined CRM known from the literature to find possible target genes. The validity of a set of putative targets is checked using Gene Ontology annotations. We demonstrate the use and effectiveness of the ModuleSearcher and ModuleScanner algorithms and test their specificity and sensitivity on semi-artificial data. Next, we search for a module in a cluster of gene expression profiles of human cell cycle genes. AVAILABILITY: The ModuleSearcher is available as a web service within the TOUCAN workbench for regulatory sequence analysis, which can be downloaded from http://www.esat.kuleuven.ac.be/~dna/BioI.

Algorithms↗

PubMed2003

Biclustering microarray data by Gibbs sampling.

MOTIVATION: Gibbs sampling has become a method of choice for the discovery of noisy patterns, known as motifs, in DNA and protein sequences. Because handling noise in microarray data presents similar challenges, we have adapted this strategy to the biclustering of discretized microarray data. RESULTS: In contrast with standard clustering that reveals genes that behave similarly over all the conditions, biclustering groups genes over only a subset of conditions for which those genes have a sharp probability distribution. We have opted for a simple probabilistic model of the biclusters because it has the key advantage of providing a transparent probabilistic interpretation of the biclusters in the form of an easily interpretable fingerprint. Furthermore, Gibbs sampling does not suffer from the problem of local minima that often characterizes Expectation-Maximization. We demonstrate the effectiveness of our approach on two synthetic data sets as well as a data set from leukemia patients.

Biomarkers, Tumor↗

PubMed2002

PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences.

PlantCARE is a database of plant cis-acting regulatory elements, enhancers and repressors. Regulatory elements are represented by positional matrices, consensus sequences and individual sites on particular promoter sequences. Links to the EMBL, TRANSFAC and MEDLINE databases are provided when available. Data about the transcription sites are extracted mainly from the literature, supplemented with an increasing number of in silico predicted data. Apart from a general description for specific transcription factor sites, levels of confidence for the experimental evidence, functional information and the position on the promoter are given as well. New features have been implemented to search for plant cis-acting regulatory elements in a query sequence. Furthermore, links are now provided to a new clustering and motif search method to investigate clusters of co-expressed genes. New regulatory elements can be sent automatically and will be added to the database after curation. The PlantCARE relational database is available via the World Wide Web at http://sphinx.rug.ac.be:8080/PlantCARE/.

Consensus Sequence↗

PubMed2002

A Gibbs sampling method to detect overrepresented motifs in the upstream regions of coexpressed genes.

Microarray experiments can reveal important information about transcriptional regulation. In our case, we look for potential promoter regulatory elements in the upstream region of coexpressed genes. Here we present two modifications of the original Gibbs sampling algorithm for motif finding (Lawrence et al., 1993). First, we introduce the use of a probability distribution to estimate the number of copies of the motif in a sequence. Second, we describe the technical aspects of the incorporation of a higher-order background model whose application we discussed in Thijs et al. (2001). Our implementation is referred to as the Motif Sampler. We successfully validate our algorithm on several data sets. First, we show results for three sets of upstream sequences containing known motifs: 1) the G-box light-response element in plants, 2) elements involved in methionine response in Saccharomyces cerevisiae, and 3) the FNR O(2)-responsive element in bacteria. We use these data sets to explain the influence of the parameters on the performance of our algorithm. Second, we show results for upstream sequences from four clusters of coexpressed genes identified in a microarray experiment on wounding in Arabidopsis thaliana. Several motifs could be matched to regulatory elements from plant defence pathways in our database of plant cis-acting regulatory elements (PlantCARE). Some other strong motifs do not have corresponding motifs in PlantCARE but are promising candidates for further analysis.

Algorithms↗

PubMed2002

INCLUSive: integrated clustering, upstream sequence retrieval and motif sampling.

INCLUSive allows automatic multistep analysis of microarray data (clustering and motif finding). The clustering algorithm (adaptive quality-based clustering) groups together genes with highly similar expression profiles. The upstream sequences of the genes belonging to a cluster are automatically retrieved from GenBank and can be fed directly into Motif Sampler, a Gibbs sampling algorithm that retrieves statistically over-represented motifs in sets of sequences, in this case upstream regions of co-expressed genes.

Algorithms↗

PubMed2002

Adaptive quality-based clustering of gene expression profiles.

MOTIVATION: Microarray experiments generate a considerable amount of data, which analyzed properly help us gain a huge amount of biologically relevant information about the global cellular behaviour. Clustering (grouping genes with similar expression profiles) is one of the first steps in data analysis of high-throughput expression measurements. A number of clustering algorithms have proved useful to make sense of such data. These classical algorithms, though useful, suffer from several drawbacks (e.g. they require the predefinition of arbitrary parameters like the number of clusters; they force every gene into a cluster despite a low correlation with other cluster members). In the following we describe a novel adaptive quality-based clustering algorithm that tackles some of these drawbacks. RESULTS: We propose a heuristic iterative two-step algorithm: First, we find in the high-dimensional representation of the data a sphere where the "density" of expression profiles is locally maximal (based on a preliminary estimate of the radius of the cluster-quality-based approach). In a second step, we derive an optimal radius of the cluster (adaptive approach) so that only the significantly coexpressed genes are included in the cluster. This estimation is achieved by fitting a model to the data using an EM-algorithm. By inferring the radius from the data itself, the biologist is freed from finding an optimal value for this radius by trial-and-error. The computational complexity of this method is approximately linear in the number of gene expression profiles in the data set. Finally, our method is successfully validated using existing data sets. AVAILABILITY: http://www.esat.kuleuven.ac.be/~thijs/Work/Clustering.html

Algorithms↗

PubMed1999

Embedding recurrent neural networks into predator-prey models.

We study changes of coordinates that allow the embedding of ordinary differential equations describing continuous-time recurrent neural networks into differential equations describing predator-prey models-also called Lotka-Volterra systems. We transform the equations for the neural network first into quasi-monomial form (Brenig, L. (1988). Complete factorization and analytic solutions of generalized Lotka-Volterra equations. Physics Letters A, 133(7-8), 378-382), where we express the vector field of the dynamical system as a linear combination of products of powers of the variables. In practice, this transformation is possible only if the activation function is the hyperbolic tangent or the logistic sigmoid. From this quasi-monomial form, we can directly transform the system further into Lotka-Volterra equations. The resulting Lotka-Volterra system is of higher dimension than the original system, but the behavior of its first variables is equivalent to the behavior of the original neural network. We expect that this transformation will permit the application of existing techniques for the analysis of Lotka-Volterra systems to recurrent neural networks. Furthermore, our results show that Lotka-Volterra systems are universal approximators of dynamical systems, just as are continuous-time neural networks.

Journal Article↗

PubMedDate not supplied

Bayesian applications of belief networks and multilayer perceptrons for ovarian tumor classification with rejection.

Incorporating prior knowledge into black-box classifiers is still much of an open problem. We propose a hybrid Bayesian methodology that consists in encoding prior knowledge in the form of a (Bayesian) belief network and then using this knowledge to estimate an informative prior for a black-box model (e.g. a multilayer perceptron). Two technical approaches are proposed for the transformation of the belief network into an informative prior. The first one consists in generating samples according to the most probable parameterization of the Bayesian belief network and using them as virtual data together with the real data in the Bayesian learning of a multilayer perceptron. The second approach consists in transforming probability distributions over belief network parameters into distributions over multilayer perceptron parameters. The essential attribute of the hybrid methodology is that it combines prior knowledge and statistical data efficiently when prior knowledge is available and the sample is of small or medium size. Additionally, we describe how the Bayesian approach can provide uncertainty information about the predictions (e.g. for classification with rejection). We demonstrate these techniques on the medical task of predicting the malignancy of ovarian masses and summarize the practical advantages of the Bayesian approach. We compare the learning curves for the hybrid methodology with those of several belief networks and multilayer perceptrons. Furthermore, we report the performance of Bayesian belief networks when they are allowed to exclude hard cases based on various measures of prediction uncertainty.

Bayes Theorem↗