Petri Törönen

Analysis of gene expression data using self-organizing maps.

DNA microarray technologies together with rapidly increasing genomic sequence information is leading to an explosion in available gene expression data. Currently there is a great need for efficient methods to analyze and visualize these massive data sets. A self‐organizing map (SOM) is an unsupervised neural network learning algorithm which has been successfully used for the analysis and organization of large data files. We have here applied the SOM algorithm to analyze published data of yeast gene expression and show that SOM is an excellent tool for the analysis and visualization of gene expression profiles.

Analysis and visualization of gene expression data using self-organizing maps

Cluster structure of gene expression data obtained from DNA microarrays is analyzed and visualized with the Self-Organizing Map (SOM) algorithm. The SOM forms a non-linear mapping of the data to a two-dimensional map grid that can be used as an exploratory data analysis tool for generating hypotheses on the relationships, and ultimately of the function of the genes. Similarity relationships within the data and cluster structures can be visualized and interpreted. The methods are demonstrated by computing a SOM of yeast genes. The relationships of known functional classes of genes are investigated by analyzing their distribution on the SOM, the cluster structure is visualized by the U-matrix method, and the clusters are characterized in terms of the properties of the expression profiles of the genes. Finally, it is shown that the SOM visualizes the similarity of genes in a more trustworthy way than two alternative methods, multidimensional scaling and hierarchical clustering.

Trustworthiness and metrics in visualizing similarity of gene expression

BackgroundConventionally, the first step in analyzing the large and high-dimensional data sets measured by microarrays is visual exploration. Dendrograms of hierarchical clustering, self-organizing maps (SOMs), and multidimensional scaling have been used to visualize similarity relationships of data samples. We address two central properties of the methods: (i) Are the visualizations trustworthy, i.e., if two samples are visualized to be similar, are they really similar? (ii) The metric. The measure of similarity determines the result; we propose using a new learning metrics principle to derive a metric from interrelationships among data sets.ResultsThe trustworthiness of hierarchical clustering, multidimensional scaling, and the self-organizing map were compared in visualizing similarity relationships among gene expression profiles. The self-organizing map was the best except that hierarchical clustering was the most trustworthy for the most similar profiles. Trustworthiness can be further increased by treating separately those genes for which the visualization is least trustworthy. We then proceed to improve the metric. The distance measure between the expression profiles is adjusted to measure differences relevant to functional classes of the genes. The genes for which the new metric is the most different from the usual correlation metric are listed and visualized with one of the visualization methods, the self-organizing map, computed in the new metric.ConclusionsThe conjecture from the methodological results is that the self-organizing map can be recommended to complement the usual hierarchical clustering for visualizing and exploring gene expression data. Discarding the least trustworthy samples and improving the metric still improves it.

Defense-related transcription factors WRKY70 and WRKY54 modulate osmotic stress tolerance by regulating stomatal aperture in Arabidopsis

WRKY transcription factors (TFs) have been mainly associated with plant defense, but recent studies have suggested additional roles in the regulation of other physiological processes. Here, we explored the possible contribution of two related group III WRKY TFs, WRKY70 and WRKY54, to osmotic stress tolerance. These TFs are positive regulators of plant defense, and co-operate as negative regulators of salicylic acid (SA) biosynthesis and senescence. We employed single and double mutants of wrky54 and wrky70, as well as a WRKY70 overexpressor line, to explore the role of these TFs in osmotic stress (polyethylene glycol) responses. Their effect on gene expression was characterized by microarrays and verified by quantitative PCR. Stomatal phenotypes were assessed by water retention and stomatal conductance measurements. The wrky54wrky70 double mutants exhibited clearly enhanced tolerance to osmotic stress. However, gene expression analysis showed reduced induction of osmotic stress-responsive genes in addition to reduced accumulation of the osmoprotectant proline. By contrast, the enhanced tolerance was correlated with improved water retention and enhanced stomatal closure. These findings demonstrate that WRKY70 and WRKY54 co-operate as negative regulators of stomatal closure and, consequently, osmotic stress tolerance in Arabidopsis, suggesting that they have an important role, not only in plant defense, but also in abiotic stress signaling.

Antipsychotic drug treatment induces differential gene expression in the rat cortex

Antipsychotic drug treatment is known to modulate gene expression in experimental animals. In this study, candidate target genes for antipsychotic drug action were searched using microarrays after acute clozapine treatment (1, 6 and 24 h) in the rat prefrontal cortex. Microarray data clustering with a self‐organizing map algorithm revealed differential expression of genes involved in presynaptic function following acute clozapine treatment. The differential expression of 35 genes most profoundly regulated in expression arrays was further examined using in situ hybridization following acute clozapine, and chronic clozapine and haloperidol treatments. Acute administration of clozapine regulated the expression of chromogranin A, synaptotagmin V and calcineurin A mRNAs in the cortex. Chronic clozapine treatment induced differential cortical expression of chromogranin A, son of sevenless (SoS) and Sec‐1. Chronic treatment with haloperidol regulated the mRNA expression of inhibitor of DNA‐binding 2 (ID‐2) and Rab‐12. Furthermore, the expression of visinin‐like proteins‐1, ‐2 and ‐3 was regulated by chronic drug treatments in various brain regions. Our data suggest that acute and chronic treatments with haloperidol and clozapine modulate the expression of genes involved in synaptic function and in regulation of intracellular Ca2+ in cortex.

Revised Phylogeny and Novel Horizontally Acquired Virulence Determinants of the Model Soft Rot Phytopathogen Pectobacterium wasabiae SCC3193

Soft rot disease is economically one of the most devastating bacterial diseases affecting plants worldwide. In this study, we present novel insights into the phylogeny and virulence of the soft rot model Pectobacterium sp. SCC3193, which was isolated from a diseased potato stem in Finland in the early 1980s. Genomic approaches, including proteome and genome comparisons of all sequenced soft rot bacteria, revealed that SCC3193, previously included in the species Pectobacterium carotovorum, can now be more accurately classified as Pectobacterium wasabiae. Together with the recently revised phylogeny of a few P. carotovorum strains and an increasing number of studies on P. wasabiae, our work indicates that P. wasabiae has been unnoticed but present in potato fields worldwide. A combination of genomic approaches and in planta experiments identified features that separate SCC3193 and other P. wasabiae strains from the rest of soft rot bacteria, such as the absence of a type III secretion system that contributes to virulence of other soft rot species. Experimentally established virulence determinants include the putative transcriptional regulator SirB, two partially redundant type VI secretion systems and two horizontally acquired clusters (Vic1 and Vic2), which contain predicted virulence genes. Genome comparison also revealed other interesting traits that may be related to life in planta or other specific environmental conditions. These traits include a predicted benzoic acid/salicylic acid carboxyl methyltransferase of eukaryotic origin. The novelties found in this work indicate that soft rot bacteria have a reservoir of unknown traits that may be utilized in the poorly understood latent stage in planta. The genomic approaches and the comparison of the model strain SCC3193 to other sequenced Pectobacterium strains, including the type strain of P. wasabiae, provides a solid basis for further investigation of the virulence, distribution and phylogeny of soft rot bacteria and, potentially, other bacteria as well.

Theme discovery from gene lists for identification and viewing of multiple functional groups.

BackgroundHigh throughput methods of the genome era produce vast amounts of data in the form of gene lists. These lists are large and difficult to interpret without advanced computational or bioinformatic tools. Most existing methods analyse a gene list as a single entity although it is comprised of multiple gene groups associated with separate biological functions. Therefore it is imperative to define and visualize gene groups with unique functionality within gene lists.ResultsIn order to analyse the functional heterogeneity within a gene list, we have developed a method that clusters genes to groups with homogenous functionalities. The method uses Non-negative Matrix Factorization (NMF) to create several clustering results with varying numbers of clusters. The obtained clustering results are combined into a simple graphical presentation showing the functional groups over-represented in the analyzed gene list. We demonstrate its performance on two data sets and show results that improve upon existing methods. The comparison also shows that our method creates a more simplified view that aids in discovery of biological themes within the list and discards less informative classes from the results.ConclusionThe presented method and associated software are useful for the identification and interpretation of biological functions associated with gene lists and are especially useful for the analysis of large lists.

Selection of informative clusters from hierarchical cluster tree with gene classes

A common clustering method in the analysis of gene expression data has been hierarchical clustering. Usually the analysis involves selection of clusters by cutting the tree at a suitable level and/or analysis of a sorted gene list that is obtained with the tree. Cutting of the hierarchical tree requires the selection of a suitable level and it results in the loss of information on the other level. Sorted gene lists depend on the sorting method of the joined clusters. Author proposes that the clusters should be selected using the gene classifications. This article presents a simple method for searching for clusters with the strongest enrichment of gene classes from a cluster tree. The clusters found are presented in the estimated order of importance. The method is demonstrated with a yeast gene expression data set and with two database classifications. The obtained clusters demonstrated a very strong enrichment of functional classes. The obtained clusters are also able to present similar gene groups to those that were observed from the data set in the original analysis and also many gene groups that were not reported in the original analysis. Visualization of the results on top of a cluster tree shows that the method finds informative clusters from several levels of the cluster tree and indicates that the clusters found could not have been obtained by simply cutting the cluster tree. Results were also used in the comparison of cluster trees from different clustering methods. The presented method should facilitate the exploratory analysis of big data sets when the associated categorical data is available.

Brain-derived neurotrophic factor signaling modifies hippocampal gene expression during epileptogenesis in transgenic mice.

Brain‐derived neurotrophic factor (BDNF) regulates neuronal survival, differentiation and plasticity. It has been shown to promote epileptogenesis and transgenic mice with decreased and increased BDNF signaling show opposite alterations in epileptogenesis. However, the mechanisms of BDNF action are largely unknown. We studied the gene expression changes 12 days after kainic acid‐induced status epilepticus in transgenic mice overexpressing either the functional BDNF receptor trkB or a dominant‐negative truncated trkB. Epileptogenesis produced marked changes in expression of 27 of 1090 genes. Cluster analysis revealed BDNF signalling‐mediated regulation of functional gene classes involved in cellular transport, DNA repair and cell death, including kinesin motor kinesin family member 3A involved in cellular transport. Furthermore, the expression of cytoskeletal and extracellular matrix components, such as tissue inhibitor of metalloproteinase 2 was altered, emphasizing the importance of intracellular transport and interplay between neurons and glia during epileptogenesis. Finally, mice overexpressing the dominant‐negative trkB, which were previously shown to have reduced epileptogenesis, showed a decrease in mRNAs of several growth‐associated genes, including growth‐associated protein 43. Our data suggest that BDNF signaling may partly mediate the development of epilepsy and propose that regrowth or repair processes initiated by status epilepticus and promoted by BDNF signaling may not be as advantageous as previously thought.

Robust extraction of functional signals from gene set analysis using a generalized threshold free scoring function

BackgroundA central task in contemporary biosciences is the identification of biological processes showing response in genome-wide differential gene expression experiments. Two types of analysis are common. Either, one generates an ordered list based on the differential expression values of the probed genes and examines the tail areas of the list for over-representation of various functional classes. Alternatively, one monitors the average differential expression level of genes belonging to a given functional class. So far these two types of method have not been combined.ResultsWe introduce a scoring function, Gene Set Z-score (GSZ), for the analysis of functional class over-representation that combines two previous analysis methods. GSZ encompasses popular functions such as correlation, hypergeometric test, Max-Mean and Random Sets as limiting cases. GSZ is stable against changes in class size as well as across different positions of the analysed gene list in tests with randomized data. GSZ shows the best overall performance in a detailed comparison to popular functions using artificial data. Likewise, GSZ stands out in a cross-validation of methods using split real data. A comparison of empirical p-values further shows a strong difference in favour of GSZ, which clearly reports better p-values for top classes than the other methods. Furthermore, GSZ detects relevant biological themes that are missed by the other methods. These observations also hold when comparing GSZ with popular program packages.ConclusionGSZ and improved versions of earlier methods are a useful contribution to the analysis of differential gene expression. The methods and supplementary material are available from the website http://ekhidna.biocenter.helsinki.fi/users/petri/public/GSZ/GSZscore.html.