Correlated Gene Expression Patterns Research Articles

New Insights on the Diurnal Mechanism of Calcification in the Stony Coral, Stylophora pistillata

Scleractinian corals are evolutionary-successful calcifying marine organisms, which utilize an endo-symbiotic relationship with photosynthetic dinoflagellate algae that supply energy products to their coral hosts. This energy further supports a higher calcification rate during the day in a process known as light enhanced calcification. Although this process has been studied for decades, the mechanisms behind it are still unknown. However, photosynthesis and respiration also cause daily fluctuations in oxygen and pH levels, resulting in the coral facing highly variable conditions. Here we correlated gene expression patterns with the physiological differences along the diel cycle to provide new insights on the daily dynamic processes, including circadian rhythm, calcification, symbiosis, cellular arrangement, metabolism, and energy budget. During daytime, when solar radiation levels are highest, we observed increased calcification rate combined with an extensive up-regulation of genes associated with reactive oxygen species, redox, metabolism, ion transporters, skeletal organic matrix, and mineral formation. During the night, we observed a vast shift toward up-regulation of genes associated with cilia movement, tissue development, cellular movement, antioxidants, protein synthesis, and skeletal organic matrix formation. Our results suggest that light enhanced calcification is related to several processes that occur across the diel cycle; during nighttime, tissue might elevate away from the skeleton, extending the calcifying space area to enable the formation of a new organic framework template. During daytime, the combination of synthesis of acid-rich proteins and a greater flux of ions to the sites of calcification facilitate the conditions for extensive mineral growth.

NIMG-44. NON-INVASIVE ADVANCED IMAGING METRICS PROVIDE POTENTIAL BIOMARKERS FOR GBM BIOLOGICAL HETEROGENEITY AND IMMUNE LANDSCAPE VARIANCE

Abstract BACKGROUND Treatment of glioblastoma multiforme (GBM) is complicated by extensive tumor heterogeneity. We hypothesize that transcriptomic analysis of brain tumor regions with different magnetic resonance imaging (MRI) characteristics will define specific biological processes, providing a non-invasive method for tumor characterization and stratification. METHODS Previously at the University of California San Francisco, treatment naïve GBM tissue from gadolinium contrast enhancing lesion (CEL) and non-enhancing lesion (NCEL) regions were stereotactically sampled; prior to resection, relative cerebral blood volume (rCBV) and apparent diffusion coefficient (ADC) were determined. The tissue samples were characterized by immunohistochemistry and assessed for gene expression by microarray analysis. We correlated gene expression patterns in the CEL, NCEL, and non-tumor gliotic brain samples with multimodal physiological imaging metrics and immunohistochemical phenotypes. Gene expression networks were probed using Gene Set Enrichment Analysis. Key immunologic genes were examined individually. RESULTS Samples with differing MRI and histological phenotypes demonstrated transcriptomic variance reflecting distinct biological networks. We found significant differences in immune pathways, with immune gene signature prominent in CEL areas, moderate in NCEL and low in gliotic non-tumor brain. Within homogenously enhancing areas of CEL and NCEL there was underlying heterogeneity detectable by variable rCBV, ADC and histological phenotypes, which correlate with differing gene expression profiles indicative of biological and immunological tumor microenvironments. Increasing rCBV was correlated with an anti-inflammatory immune response in the CEL and a pro-inflammatory immune response in the NCEL. ADC was negatively correlated with cell cycle and immune networks in CEL, while NCEL ADC was positively correlated with immune processes. GBM samples with the mesenchymal molecular subtype had the greatest immune response. CONCLUSIONS Multimodal MRI features identify regionally diverse transcriptomic-based biological and immunological phenotypes in GBM. We propose that imaging genomics provides a technique for localizing biological processes and tumor immune microenvironments across space and time in GBM.

Abstract 1928: High-grade serous ovarian cancer subtypes are similar across diverse populations

Abstract The most common and lethal type of invasive epithelial ovarian cancer is high grade serous (HGSC). Three to four gene expression-based HGSC subtypes have been identified in prior studies. In contrast to most previous studies, which have assessed the performance of survival classifiers in validation sets, we sought to determine the degree of similarity of gene expression patterns in subtypes between populations using systematic unsupervised clustering within populations. We analyzed publically-available mRNA expression data from studies with >200 HGSC tumors: The Cancer Genome Atlas (TCGA, US, n = 519, Affymetrix HT U133a), Tothill et al. (GSE9891, Australia, n = 242, Affymetrix U133 Plus 2.0) and Yoshihara et al. (GSE32062, Japan, n = 258, Agilent G4112a). We restricted analyses to the 12,249 genes shared across all datasets and selected from these the union of the 1,500 most variant genes per population (2,824). Using these datasets, we performed k-means clustering within each population for k = 3 and k = 4. We compared each cluster to all other clusters using Significance Analysis of Microarrays, which outputs an F score for all 12,249 genes, measuring cluster-specific differential expression. We then calculated the correlation of the resulting F score vectors across populations and within populations across both numbers of centroids (k = 3 or k = 4). We identified analogous clusters by high F score correlations and determined each cluster's similarity to the TCGA subtypes based on cluster-specific differentially expressed genes. We observed high concordance of gene expression patterns for clusters across populations and across k-means runs, suggesting that analogous clusters exist in most analyses. For k = 3, F score correlations across populations for clusters 1, 2 and 3, respectively, ranged between 0.77-0.85, 0.80-0.90, and 0.66-0.72. For k = 4, F score correlations for clusters 1-4 were, respectively: 0.76-0.85, 0.82-0.85, 0.65-0.78, and 0.52-0.78. Across k = 3 and k = 4, correlations for cluster 1 within TCGA, Tothill, and Yoshihara were 0.99, 1.00 and 1.00, and correlations for cluster 2 were 0.96, 0.98, and 0.95, respectively. Correlations for cluster 3 were less strong: 0.56, 0.88, and 0.60, respectively. For k = 4, cluster 4 was composed mainly of samples that belonged to cluster 3 for k = 3; 88% for TCGA, 54% for Tothill, and 95% for Yoshihara. When compared to TCGA subtypes, cluster 1 corresponded most strongly to mesenchymal, cluster 2 to proliferative, cluster 3 to differentiated, and cluster 4 to immunoreactive. Our observation of highly correlated gene expression patterns between clusters across populations, across platforms, and across the number of centroids provides strong evidence that at least three biological HGSC subtypes exist. The mesenchymal-like and proliferative-like subtypes are particularly consistent across populations, and could be uniquely targeted for treatment. Citation Format: Gregory P. Way, James Rudd, Casey S. Greene, Jennifer A. Doherty. High-grade serous ovarian cancer subtypes are similar across diverse populations. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1928. doi:10.1158/1538-7445.AM2015-1928

Using gene expression signatures to identify novel treatment strategies in gulf war illness.

BackgroundGulf War Illness (GWI) is a complex multi-symptom disorder that affects up to one in three veterans of this 1991 conflict and for which no effective treatment has been found. Discovering novel treatment strategies for such a complex chronic illness is extremely expensive, carries a high probability of failure and a lengthy cycle time. Repurposing Food and Drug Administration approved drugs offers a cost-effective solution with a significantly abbreviated timeline.MethodsHere, we explore drug re-purposing opportunities in GWI by combining systems biology and bioinformatics techniques with pharmacogenomic information to find overlapping elements in gene expression linking GWI to successfully treated diseases. Gene modules were defined based on cellular function and their activation estimated from the differential expression of each module’s constituent genes. These gene modules were then cross-referenced with drug atlas and pharmacogenomic databases to identify agents currently used successfully for treatment in other diseases. To explore the clinical use of these drugs in illnesses similar to GWI we compared gene expression patterns in modules that were significantly expressed in GWI with expression patterns in those same modules in other illnesses.ResultsWe found 19 functional modules with significantly altered gene expression patterns in GWI. Within these modules, 45 genes were documented drug targets. Illnesses with highly correlated gene expression patterns overlapping considerably with GWI were found in 18 of the disease conditions studied. Brain, muscular and autoimmune disorders composed the bulk of these.ConclusionOf the associated drugs, immunosuppressants currently used in treating rheumatoid arthritis, and hormone based therapies were identified as the best available candidates for treating GWI symptoms.Electronic supplementary materialThe online version of this article (doi:10.1186/s12920-015-0111-3) contains supplementary material, which is available to authorized users.

Microarray Analysis in Glioblastomas.

Microarray analysis in glioblastomas is done using either cell lines or patient samples as starting material. A survey of the current literature points to transcript-based microarrays and immunohistochemistry (IHC)-based tissue microarrays as being the preferred methods of choice in cancers of neurological origin. Microarray analysis may be carried out for various purposes including the following: i. To correlate gene expression signatures of glioblastoma cell lines or tumors with response to chemotherapy (DeLay et al., Clin Cancer Res 18(10):2930-2942, 2012). ii. To correlate gene expression patterns with biological features like proliferation or invasiveness of the glioblastoma cells (Jiang et al., PLoS One 8(6):e66008, 2013). iii. To discover new tumor classificatory systems based on gene expression signature, and to correlate therapeutic response and prognosis with these signatures (Huse et al., Annu Rev Med 64(1):59-70, 2013; Verhaak et al., Cancer Cell 17(1):98-110, 2010). While investigators can sometimes use archived tumor gene expression data available from repositories such as the NCBI Gene Expression Omnibus to answer their questions, new arrays must often be run to adequately answer specific questions. Here, we provide a detailed description of microarray methodologies, how to select the appropriate methodology for a given question, and analytical strategies that can be used. Experimental methodology for protein microarrays is outside the scope of this chapter, but basic sample preparation techniques for transcript-based microarrays are included here.

The expansion of body coloration involves coordinated evolution in cis and trans within the pigmentation regulatory network of Drosophila prostipennis

The generation of complex morphological features requires the precisely orchestrated expression of numerous genes during development. While several traits have been resolved to evolutionary changes within a single gene, the evolutionary path by which genes derive co-localized or mutually excluded expression patterns is currently a mystery. Here we investigate how the Drosophila pigmentation gene network was altered in Drosophila prostipennis, a species in the Drosophila melanogaster subgroup, that evolved expanded abdominal pigmentation. We show that this expansion involved broadened expression of the melanin-promoting enzyme genes tan and yellow, and a reciprocal withdrawn pattern of the melanin-suppressing enzyme gene ebony. To examine whether these coordinated changes to the network were generated through mutations in the cis-regulatory elements (CREs) of these genes, we cloned and tested CREs of D. prostipennis tan, ebony, and yellow in transgenic reporter assays. Regulatory regions of both tan and ebony failed to recapitulate the derived D. prostipennis expression phenotype, implicating the modification of a factor or factors upstream of both genes. However, the D. prostipennis yellow cis-regulatory region recapitulated the expanded expression pattern observed in this species, implicating causative mutations in cis to yellow. Our results provide an example in which a coordinated expression program evolved through independent changes at multiple loci, rather than through changes to a single “master regulator” directing a suite of downstream target genes. This implies a complex network structure in which each gene may be subject to a unique set of inputs, and resultantly may require individualized evolutionary paths to yield correlated gene expression patterns.

Gene Regulation, Modulation, and Their Applications in Gene Expression Data Analysis

Common microarray and next-generation sequencing data analysis concentrate on tumor subtype classification, marker detection, and transcriptional regulation discovery during biological processes by exploring the correlated gene expression patterns and their shared functions. Genetic regulatory network (GRN) based approaches have been employed in many large studies in order to scrutinize for dysregulation and potential treatment controls. In addition to gene regulation and network construction, the concept of the network modulator that has significant systemic impact has been proposed, and detection algorithms have been developed in past years. Here we provide a unified mathematic description of these methods, followed with a brief survey of these modulator identification algorithms. As an early attempt to extend the concept to new RNA regulation mechanism, competitive endogenous RNA (ceRNA), into a modulator framework, we provide two applications to illustrate the network construction, modulation effect, and the preliminary finding from these networks. Those methods we surveyed and developed are used to dissect the regulated network under different modulators. Not limit to these, the concept of “modulation” can adapt to various biological mechanisms to discover the novel gene regulation mechanisms.

Pattern and polarity in the development and evolution of the gnathostome jaw: Both conservation and heterotopy in the branchial arches of the shark, Scyliorhinus canicula

The acquisition of jaws constitutes a landmark event in vertebrate evolution, one that in large part potentiated their success and diversification. Jaw development and patterning involves an intricate spatiotemporal series of reciprocal inductive and responsive interactions between the cephalic epithelia and the cranial neural crest (CNC) and cephalic mesodermal mesenchyme. The coordinated regulation of these interactions is critical for both the ontogenetic registration of the jaws and the evolutionary elaboration of variable jaw morphologies and designs. Current models of jaw development and evolution have been built on molecular and cellular evidence gathered mostly in amniotes such as mice, chicks and humans, and augmented by a much smaller body of work on the zebrafish. These have been partnered by essential work attempting to understand the origins of jaws that has focused on the jawless lamprey. Chondrichthyans (cartilaginous fish) are the most distant group to amniotes within extant gnathostomes, and comprise the crucial clade uniting amniotes and agnathans; yet despite their critical phylogenetic position, evidence of the molecular and cellular underpinnings of jaw development in chondrichthyans is still lacking. Recent advances in genome and molecular developmental biology of the lesser spotted dogfish shark, Scyliorhinus canicula, make it ideal for the molecular study of chondrichthyan jaw development. Here, following the ‘Hinge and Caps' model of jaw development, we have investigated evidence of heterotopic (relative changes in position) and heterochronic (relative changes in timing) shifts in gene expression, relative to amniotes, in the jaw primordia of S. canicula embryos. We demonstrate the presence of clear proximo-distal polarity in gene expression patterns in the shark embryo, thus establishing a baseline molecular baüplan for branchial arch-derived jaw development and further validating the utility of the ‘Hinge and Caps’ model in comparative studies of jaw development and evolution. Moreover, we correlate gene expression patterns with the absence of a lambdoidal junction (formed where the maxillary first arch meets the frontonasal processes) in chondrichthyans, further highlighting the importance of this region for the development and evolution of jaw structure in advanced gnathostomes.

Biclustering for the comprehensive search of correlated gene expression patterns using clustered seed expansion.

BackgroundIn a functional analysis of gene expression data, biclustering method can give crucial information by showing correlated gene expression patterns under a subset of conditions. However, conventional biclustering algorithms still have some limitations to show comprehensive and stable outputs.ResultsWe propose a novel biclustering approach called “BIclustering by Correlated and Large number of Individual Clustered seeds (BICLIC)” to find comprehensive sets of correlated expression patterns in biclusters using clustered seeds and their expansion with correlation of gene expression. BICLIC outperformed competing biclustering algorithms by completely recovering implanted biclusters in simulated datasets with various types of correlated patterns: shifting, scaling, and shifting-scaling. Furthermore, in a real yeast microarray dataset and a lung cancer microarray dataset, BICLIC found more comprehensive sets of biclusters that are significantly enriched to more diverse sets of biological terms than those of other competing biclustering algorithms.ConclusionsBICLIC provides significant benefits in finding comprehensive sets of correlated patterns and their functional implications from a gene expression dataset.

Characterization of Circadian Behavior in the Starlet Sea Anemone, Nematostella vectensis

BackgroundAlthough much is known about how circadian systems control daily cycles in the physiology and behavior of Drosophila and several vertebrate models, marine invertebrates have often been overlooked in circadian rhythms research. This study focuses on the starlet sea anemone, Nematostella vectensis, a species that has received increasing attention within the scientific community for its potential as a model research organism. The recently sequenced genome of N. vectensis makes it an especially attractive model for exploring the molecular evolution of circadian behavior. Critical behavioral data needed to correlate gene expression patterns to specific behaviors are currently lacking in N. vectensis.Methodology/Principal FindingsTo detect the presence of behavioral oscillations in N. vectensis, locomotor activity was evaluated using an automated system in an environmentally controlled chamber. Animals exposed to a 24 hr photoperiod (12 hr light: 12 hr dark) exhibited locomotor behavior that was both rhythmic and predominantly nocturnal. The activity peak occurred in the early half of the night with a 2-fold increase in locomotion. Upon transfer to constant lighting conditions (constant light or constant dark), an approximately 24 hr rhythm persisted in most animals, suggesting that the rhythm is controlled by an endogenous circadian mechanism. Fourier analysis revealed the presence of multiple peaks in some animals suggesting additional rhythmic components could be present. In particular, an approximately 12 hr oscillation was often observed. The nocturnal increase in generalized locomotion corresponded to a 24 hr oscillation in animal elongation.Conclusions/SignificanceThese data confirm the presence of a light-entrainable circadian clock in Nematostella vectensis. Additional components observed in some individuals indicate that an endogenous clock of approximately 12 hr frequency may also be present. By describing rhythmic locomotor behavior in N. vectensis, we have made important progress in developing the sea anemone as a model organism for circadian rhythm research.

Comparative co‐expression analysis in plant biology

The analysis of gene expression data generated by high-throughput microarray transcript profiling experiments has shown that transcriptionally coordinated genes are often functionally related. Based on large-scale expression compendia grouping multiple experiments, this guilt-by-association principle has been applied to study modular gene programmes, identify cis-regulatory elements or predict functions for unknown genes in different model plants. Recently, several studies have demonstrated how, through the integration of gene homology and expression information, correlated gene expression patterns can be compared between species. The incorporation of detailed functional annotations as well as experimental data describing protein-protein interactions, phenotypes or tissue specific expression, provides an invaluable source of information to identify conserved gene modules and translate biological knowledge from model organisms to crops. In this review, we describe the different steps required to systematically compare expression data across species. Apart from the technical challenges to compute and display expression networks from multiple species, some future applications of plant comparative transcriptomics are highlighted.

Functional genome annotation ofDrosophilaseminal fluid proteins using transcriptional genetic networks

Predicting functional gene annotations remains a significant challenge, even in well-annotated genomes such as yeast and Drosophila. One promising, high-throughput method for gene annotation is to use correlated gene expression patterns to annotate target genes based on the known function of focal genes. The Drosophila melanogaster transcriptome varies genetically among wild-derived inbred lines, with strong genetic correlations among the transcripts. Here, we leveraged the genetic correlations in gene expression among known seminal fluid protein (SFP) genes and the rest of the genetically varying transcriptome to identify 176 novel candidate SFPs (cSFPs). We independently validated the correlation in gene expression between seven of the cSFPs and a known SFP gene, as well as expression in male reproductive tissues. We argue that this method can be extended to other systems for which information on genetic variation in gene expression is available.

Comparative Network Analysis Reveals That Tissue Specificity and Gene Function Are Important Factors Influencing the Mode of Expression Evolution in Arabidopsis and Rice

Microarray experiments have yielded massive amounts of expression information measured under various conditions for the model species Arabidopsis (Arabidopsis thaliana) and rice (Oryza sativa). Expression compendia grouping multiple experiments make it possible to define correlated gene expression patterns within one species and to study how expression has evolved between species. We developed a robust framework to measure expression context conservation (ECC) and found, by analyzing 4,630 pairs of orthologous Arabidopsis and rice genes, that 77% showed conserved coexpression. Examples of nonconserved ECC categories suggested a link between regulatory evolution and environmental adaptations and included genes involved in signal transduction, response to different abiotic stresses, and hormone stimuli. To identify genomic features that influence expression evolution, we analyzed the relationship between ECC, tissue specificity, and protein evolution. Tissue-specific genes showed higher expression conservation compared with broadly expressed genes but were fast evolving at the protein level. No significant correlation was found between protein and expression evolution, implying that both modes of gene evolution are not strongly coupled in plants. By integration of cis-regulatory elements, many ECC conserved genes were significantly enriched for shared DNA motifs, hinting at the conservation of ancestral regulatory interactions in both model species. Surprisingly, for several tissue-specific genes, patterns of concerted network evolution were observed, unveiling conserved coexpression in the absence of conservation of tissue specificity. These findings demonstrate that orthologs inferred through sequence similarity in many cases do not share similar biological functions and highlight the importance of incorporating expression information when comparing genes across species.

Upregulation of the 5-Lipoxygenase Pathway in Human Aortic Valves Correlates With Severity of Stenosis and Leads to Leukotriene-Induced Effects on Valvular Myofibroblasts

The development of aortic valve stenosis is not only associated with calcification and extracellular matrix remodeling, but also with inflammation. The aim of this study was to determine the role of proinflammatory signaling through the leukotriene (LT) pathway in aortic stenosis. After macroscopic dissection of surgically removed human aortic valves, RNA was extracted from 311 preparations derived from 68 patients to differentiate normal, thickened, and calcified areas from each cusp. Subsequently, quantitative polymerase chain reaction analysis was used to correlate gene expression patterns with preoperative echocardiographic parameters. The messenger RNA levels of the LT-forming enzyme 5-lipoxygenase increased 1.6- and 2.2-fold in thickened and calcified tissue, respectively, compared with normal areas of the same valves. In thickened tissues, messenger RNA levels for 5-lipoxygenase (r= -0.35; P=0.03), its activating protein (5-lipoxygenase activating protein; r= -0.39; P=0.02), and LTA(4) hydrolase (r= -0.48; P=0.01) correlated inversely with the velocity-time integral ratio. In addition, leukotriene A(4) hydrolase transcripts correlated inversely with aortic valve area, indexed for body surface area (r= -0.52; P=0.007). Immunohistochemical stainings revealed LT receptor expression on valvular myofibroblasts. In primary cultures of human myofibroblasts derived from stenotic aortic valves, Leukotriene C(4) (LTC(4)) increased intracellular calcium, enhanced reactive oxygen species production, reduced the mitochondrial membrane potential, and led to morphological cell cytoplasm changes and calcification. The upregulation of the LT pathway in human aortic valve stenosis and its correlation with clinical stenosis severity, taken together with the potentially detrimental LT-induced effects on valvular myofibroblasts, suggests one possible role of inflammation in the development of aortic stenosis.

Gene Expression Profiling for Survival Prediction in Pediatric Rhabdomyosarcomas: A Report From the Children's Oncology Group

We investigated whether tumors from diagnostic biopsies of primary rhabdomyosarcoma (RMS) contain relevant prognostic information in the form of gene expression signatures that can be used to model and predict outcome of patients. A 22,000-probe set microarray was used to evaluate 120 RMS specimens and correlate gene expression patterns to survival. Multivariate gene expression models or metagenes were developed using cross-validated Cox regression proportional hazards modeling and were evaluated using Kaplan-Meier analysis. A 34-metagene, based on expression patterns of 34 genes, was highly predictive of outcome. It was not highly correlated with individual clinical risk factors such as patient age, stage, tumor size, or histology. However, it was correlated with a risk classification used by the Children's Oncology Group and the biologic subsets of alveolar histology tumors. These data support further evaluation of RMS metagenes to discriminate patients with good prognosis from those with poor prognosis, with the potential to direct risk-adapted therapy.

Gene Expression Profiles During Hepatic Stellate Cell Activation in Culture and In Vivo

Following hepatic injury, hepatic stellate cells (HSCs) transdifferentiate to become extracellular matrix-producing myofibroblasts and to promote hepatic fibrogenesis. In this study, we determine gene expression changes in 3 different models of HSC activation and investigate whether HSC culture activation reproduces gene expression changes of HSC in vivo activation. HSCs were isolated by density centrifugation and magnetic antibody cell sorting from normal mice, CCl(4)-treated mice, and mice that underwent bile duct ligation (BDL). Gene expression was analyzed by microarray and confirmed by polymerase chain reaction and Western blot analysis. Two thousand seventy-three probe sets were differentially expressed in at least 1 of 3 models of HSC activation, including novel genes that encode proinflammatory and antiapoptotic mediators; transcription factors; cell surface receptors; and cytoskeleton components such as CXCL14, survivin, septin 4, osteopontin, PRX1, LMCD1, GPR91, leiomodin, and anillin. BDL- and CCl(4)-activated HSCs showed highly correlated gene expression patterns, whereas culture activation only partially reproduced the gene expression changes observed during BDL- and CCl(4)-induced activation. Coculture with Kupffer cells or lipopolysaccharide treatment during culture activation shifted the expression of most examined genes toward the pattern observed during in vivo activation, suggesting a role for these factors in the microenvironment that drives HSC activation. The almost identical HSC gene expression patterns after BDL or CCl(4) treatment indicate that HSCs exert similar functions in different types of liver injury. Because culture activation does not properly regulate gene expression in HSCs, in vivo activation should be considered the gold standard for the study of HSC biology.

Real-time gene expression analysis in carp ( Cyprinus carpio L.) skin: Inflammatory responses to injury mimicking infection with ectoparasites

We studied a predictive model of gene expression induced by mechanical injury of fish skin, to resolve the confounding effects on the immune system induced by injury and skin parasite-specific molecules. We applied real time quantitative PCR (RQ-PCR) to measure the expression of the pro-inflammatory cytokines CXCa, CXCb, interleukin (IL)1- β, tumor necrosis factor α (TNF α), and the receptors IL1R1, CXCR1 and CXCR2 in skin of Cyprinus carpio after mechanical injury. We also studied the expression of the anti-inflammatory cytokine IL-10. Most obvious, specific up-regulation of the chemokine CXCa, the chemokine receptor CXCR1 and the pro-inflammatory cytokine IL- β was detected at 2–3 h after injury. In order to correlate gene expression patterns after injury with cell migration, we studied chemotaxis of head kidney leukocytes towards lysates of epithelioma papulosum cyprini (EPC) cells. Neutrophilic granulocytes were shown to migrate towards epithelial lysates. Using immunohistochemistry we observed that the early inflammatory response after injury involved an influx of cells, most probably neutrophilic granulocytes, into the injured area. This suggests that the increased expression of pro-inflammatory genes is related to a rapid influx of neutrophilic granulocytes.

A hierarchical unsupervised growing neural network for clustering gene expression patterns.

We describe a new approach to the analysis of gene expression data coming from DNA array experiments, using an unsupervised neural network. DNA array technologies allow monitoring thousands of genes rapidly and efficiently. One of the interests of these studies is the search for correlated gene expression patterns, and this is usually achieved by clustering them. The Self-Organising Tree Algorithm, (SOTA) (Dopazo,J. and Carazo,J.M. (1997) J. Mol. Evol., 44, 226-233), is a neural network that grows adopting the topology of a binary tree. The result of the algorithm is a hierarchical cluster obtained with the accuracy and robustness of a neural network. SOTA clustering confers several advantages over classical hierarchical clustering methods. SOTA is a divisive method: the clustering process is performed from top to bottom, i.e. the highest hierarchical levels are resolved before going to the details of the lowest levels. The growing can be stopped at the desired hierarchical level. Moreover, a criterion to stop the growing of the tree, based on the approximate distribution of probability obtained by randomisation of the original data set, is provided. By means of this criterion, a statistical support for the definition of clusters is proposed. In addition, obtaining average gene expression patterns is a built-in feature of the algorithm. Different neurons defining the different hierarchical levels represent the averages of the gene expression patterns contained in the clusters. Since SOTA runtimes are approximately linear with the number of items to be classified, it is especially suitable for dealing with huge amounts of data. The method proposed is very general and applies to any data providing that they can be coded as a series of numbers and that a computable measure of similarity between data items can be used. A server running the program can be found at: http://bioinfo.cnio.es/sotarray.