Application of text mining in the biomedical domain.

In recent years the amount of experimental data that is produced in biomedical research and the number of papers that are being published in this field have grown rapidly. In order to keep up to date with developments in their field of interest and to interpret the outcome of experiments in light of all available literature, researchers turn more and more to the use of automated literature mining. As a consequence, text mining tools have evolved considerably in number and quality and nowadays can be used to address a variety of research questions ranging from de novo drug target discovery to enhanced biological interpretation of the results from high throughput experiments. In this paper we introduce the most important techniques that are used for a text mining and give an overview of the text mining tools that are currently being used and the type of problems they are typically applied for.

Identification of gene signatures for prednisolone-induced metabolic dysfunction in collagen-induced arthritic mice.

BACKGROUND: Prednisolone is a potent anti-inflammatory glucocorticoid (GC) but chronic use is hampered by metabolic side effects. Little is known about the long-term effects of GCs on gene-expression in vivo during inflammation. AIM: Identify gene signatures underlying prednisolone-induced metabolic side effects in a complex in vivo inflammatory setting after long-term treatment. MATERIALS & METHODS: We performed whole-genome expression profiling in liver and muscle from arthritic and nonarthritic mice treated with several doses of prednisolone for 3 weeks and used text-mining to link gene signatures to metabolic pathways. RESULTS: Prednisolone-induced gene signatures were highly tissue specific. We identified a short-list of genes significantly affected by both prednisolone and inflammation in liver and involved in glucose and fatty acid metabolism. For several of these genes the association with GCs is novel. CONCLUSION: The identified gene signatures may provide useful starting points for the development of GCs with a better safety profile.

Prednisolone induces the Wnt signalling pathway in 3T3-L1 adipocytes.

Synthetic glucocorticoids are potent anti-inflammatory drugs but show dose-dependent metabolic side effects such as the development of insulin resistance and obesity. The precise mechanisms involved in these glucocorticoid-induced side effects, and especially the participation of adipose tissue in this are not completely understood. We used a combination of transcriptomics, antibody arrays and bioinformatics approaches to characterize prednisolone-induced alterations in gene expression and adipokine secretion, which could underlie metabolic dysfunction in 3T3-L1 adipocytes. Several pathways, including cytokine signalling, Akt signalling, and Wnt signalling were found to be regulated at multiple levels, showing that these processes are targeted by prednisolone. These results suggest that mechanisms by which prednisolone induce insulin resistance include dysregulation of wnt signalling and immune response processes. These pathways may provide interesting targets for the development of improved glucocorticoids.

Identification of new biomarker candidates for glucocorticoid induced insulin resistance using literature mining.

BACKGROUND: Glucocorticoids are potent anti-inflammatory agents used for the treatment of diseases such as rheumatoid arthritis, asthma, inflammatory bowel disease and psoriasis. Unfortunately, usage is limited because of metabolic side-effects, e.g. insulin resistance, glucose intolerance and diabetes. To gain more insight into the mechanisms behind glucocorticoid induced insulin resistance, it is important to understand which genes play a role in the development of insulin resistance and which genes are affected by glucocorticoids.Medline abstracts contain many studies about insulin resistance and the molecular effects of glucocorticoids and thus are a good resource to study these effects. RESULTS: We developed CoPubGene a method to automatically identify gene-disease associations in Medline abstracts. We used this method to create a literature network of genes related to insulin resistance and to evaluate the importance of the genes in this network for glucocorticoid induced metabolic side effects and anti-inflammatory processes.With this approach we found several genes that already are considered markers of GC induced IR, such as phosphoenolpyruvate carboxykinase (PCK) and glucose-6-phosphatase, catalytic subunit (G6PC). In addition, we found genes involved in steroid synthesis that have not yet been recognized as mediators of GC induced IR. CONCLUSIONS: With this approach we are able to construct a robust informative literature network of insulin resistance related genes that gave new insights to better understand the mechanisms behind GC induced IR. The method has been set up in a generic way so it can be applied to a wide variety of disease networks.

Molecular pathway profiling of T lymphocyte signal transduction pathways; Th1 and Th2 genomic fingerprints are defined by TCR and CD28-mediated signaling.

BACKGROUND: T lymphocytes are orchestrators of adaptive immunity. Naïve T cells may differentiate into Th1, Th2, Th17 or iTreg phenotypes, depending on environmental co-stimulatory signals. To identify genes and pathways involved in differentiation of Jurkat T cells towards Th1 and Th2 subtypes we performed comprehensive transcriptome analyses of Jurkat T cells stimulated with various stimuli and pathway inhibitors. Results from these experiments were validated in a human experimental setting using whole blood and purified CD4+ Tcells. RESULTS: Calcium-dependent activation of T cells using CD3/CD28 and PMA/CD3 stimulation induced a Th1 expression profile reflected by increased expression of T-bet, RUNX3, IL-2, and IFNγ, whereas calcium-independent activation via PMA/CD28 induced a Th2 expression profile which included GATA3, RXRA, CCL1 and Itk. Knock down with siRNA and gene expression profiling in the presence of selective kinase inhibitors showed that proximal kinases Lck and PKCθ are crucial signaling hubs during T helper cell activation, revealing a clear role for Lck in Th1 development and for PKCθ in both Th1 and Th2 development. Medial signaling via MAPkinases appeared to be less important in these pathways, since specific inhibitors of these kinases displayed a minor effect on gene expression. Translation towards a primary, whole blood setting and purified human CD4+ T cells revealed that PMA/CD3 stimulation induced a more pronounced Th1 specific, Lck and PKCθ dependent IFNγ production, whereas PMA/CD28 induced Th2 specific IL-5 and IL-13 production, independent of Lck activation. PMA/CD3-mediated skewing towards a Th1 phenotype was also reflected in mRNA expression of the master transcription factor Tbet, whereas PMA/CD28-mediated stimulation enhanced GATA3 mRNA expression in primary human CD4+ Tcells. CONCLUSIONS: This study identifies stimulatory pathways and gene expression profiles for in vitro skewing of T helper cell activation. PMA/CD3 stimulation enhances a Th1-like response in an Lck and PKCθ dependent fashion, whereas PMA/CD28 stimulation results in a Th2-like phenotype independent of the proximal TCR-tyrosine kinase Lck. This approach offers a robust and fast translational in vitro system for skewed T helper cell responses in Jurkat T cells, primary human CD4+ Tcells and in a more complex matrix such as human whole blood.

ss-TEA: Entropy based identification of receptor specific ligand binding residues from a multiple sequence alignment of class A GPCRs.

BACKGROUND: G-protein coupled receptors (GPCRs) are involved in many different physiological processes and their function can be modulated by small molecules which bind in the transmembrane (TM) domain. Because of their structural and sequence conservation, the TM domains are often used in bioinformatics approaches to first create a multiple sequence alignment (MSA) and subsequently identify ligand binding positions. So far methods have been developed to predict the common ligand binding residue positions for class A GPCRs. RESULTS: Here we present 1) ss-TEA, a method to identify specific ligand binding residue positions for any receptor, predicated on high quality sequence information. 2) The largest MSA of class A non olfactory GPCRs in the public domain consisting of 13324 sequences covering most of the species homologues of the human set of GPCRs. A set of ligand binding residue positions extracted from literature of 10 different receptors shows that our method has the best ligand binding residue prediction for 9 of these 10 receptors compared to another state-of-the-art method. CONCLUSIONS: The combination of the large multi species alignment and the newly introduced residue selection method ss-TEA can be used to rapidly identify subfamily specific ligand binding residues. This approach can aid the design of site directed mutagenesis experiments, explain receptor function and improve modelling. The method is also available online via GPCRDB at http://www.gpcr.org/7tm/.

Prednisolone-induced changes in gene-expression profiles in healthy volunteers.

BACKGROUND: Prednisolone and other glucocorticoids (GCs) are potent anti-inflammatory and immunosuppressive drugs. However, prolonged use at a medium or high dose is hampered by side effects of which the metabolic side effects are most evident. Relatively little is known about their effect on gene-expression in vivo, the effect on cell subpopulations and the relation to the efficacy and side effects of GCs. AIM: To identify and compare prednisolone-induced gene signatures in CD4⁺ T lymphocytes and CD14⁺ monocytes derived from healthy volunteers and to link these signatures to underlying biological pathways involved in metabolic adverse effects. MATERIALS & METHODS: Whole-genome expression profiling was performed on CD4⁺ T lymphocytes and CD14⁺ monocytes derived from healthy volunteers treated with prednisolone. Text-mining analyses was used to link genes to pathways involved in metabolic adverse events. RESULTS: Induction of gene-expression was much stronger in CD4⁺ T lymphocytes than in CD14⁺ monocytes with respect to fold changes, but the number of truly cell-specific genes where a strong prednisolone effect in one cell type was accompanied by a total lack of prednisolone effect in the other cell type, was relatively low. Subsequently, a large set of genes was identified with a strong link to metabolic processes, for some of which the association with GCs is novel. CONCLUSION: The identified gene signatures provide new starting points for further study into GC-induced transcriptional regulation in vivo and the mechanisms underlying GC-mediated metabolic side effects.

CoPub update: CoPub 5.0 a text mining system to answer biological questions.

In this article, we present CoPub 5.0, a publicly available text mining system, which uses Medline abstracts to calculate robust statistics for keyword co-occurrences. CoPub was initially developed for the analysis of microarray data, but we broadened the scope by implementing new technology and new thesauri. In CoPub 5.0, we integrated existing CoPub technology with new features, and provided a new advanced interface, which can be used to answer a variety of biological questions. CoPub 5.0 allows searching for keywords of interest and its relations to curated thesauri and provides highlighting and sorting mechanisms, using its statistics, to retrieve the most important abstracts in which the terms co-occur. It also provides a way to search for indirect relations between genes, drugs, pathways and diseases, following an ABC principle, in which A and C have no direct connection but are connected via shared B intermediates. With CoPub 5.0, it is possible to create, annotate and analyze networks using the layout and highlight options of Cytoscape web, allowing for literature based systems biology. Finally, operations of the CoPub 5.0 Web service enable to implement the CoPub technology in bioinformatics workflows. CoPub 5.0 can be accessed through the CoPub portal http://www.copub.org.

Prednisolone-induced differential gene expression in mouse liver carrying wild type or a dimerization-defective glucocorticoid receptor.

BACKGROUND: Glucocorticoids (GCs) control expression of a large number of genes via binding to the GC receptor (GR). Transcription may be regulated either by binding of the GR dimer to DNA regulatory elements or by protein-protein interactions of GR monomers with other transcription factors. Although the type of regulation for a number of individual target genes is known, the relative contribution of both mechanisms to the regulation of the entire transcriptional program remains elusive. To study the importance of GR dimerization in the regulation of gene expression, we performed gene expression profiling of livers of prednisolone-treated wild type (WT) and mice that have lost the ability to form GR dimers (GRdim). RESULTS: The GR target genes identified in WT mice were predominantly related to glucose metabolism, the cell cycle, apoptosis and inflammation. In GRdim mice, the level of prednisolone-induced gene expression was significantly reduced compared to WT, but not completely absent. Interestingly, for a set of genes, involved in cell cycle and apoptosis processes and strongly related to Foxo3a and p53, induction by prednisolone was completely abolished in GRdim mice. In contrast, glucose metabolism-related genes were still modestly upregulated in GRdim mice upon prednisolone treatment. Finally, we identified several novel GC-inducible genes from which Fam107a, a putative histone acetyltransferase complex interacting protein, was most strongly dependent on GR dimerization. CONCLUSIONS: This study on prednisolone-induced effects in livers of WT and GRdim mice identified a number of interesting candidate genes and pathways regulated by GR dimers and sheds new light onto the complex transcriptional regulation of liver function by GCs.

A family-based approach reveals the function of residues in the nuclear receptor ligand-binding domain.

Literature studies, 3D structure data, and a series of sequence analysis techniques were combined to reveal important residues in the structure and function of the ligand-binding domain of nuclear hormone receptors. A structure-based multiple sequence alignment allowed for the seamless combination of data from many different studies on different receptors into one single functional model. It was recently shown that a combined analysis of sequence entropy and variability can divide residues in five classes; (1) the main function or active site, (2) support for the main function, (3) signal transduction, (4) modulator or ligand binding and (5) the rest. Mutation data extracted from the literature and intermolecular contacts observed in nuclear receptor structures were analyzed in view of this classification and showed that the main function or active site residues of the nuclear receptor ligand-binding domain are involved in cofactor recruitment. Furthermore, the sequence entropy-variability analysis identified the presence of signal transduction residues that are located between the ligand, cofactor and dimer sites, suggesting communication between these regulatory binding sites. Experimental and computational results agreed well for most residues for which mutation data and intermolecular contact data were available. This allows us to predict the role of the residues for which no functional data is available yet. This study illustrates the power of family-based approaches towards the analysis of protein function, and it points out the problems and possibilities presented by the massive amounts of data that are becoming available in the "omics era". The results shed light on the nuclear receptor family that is involved in processes ranging from cancer to infertility, and that is one of the more important targets in the pharmaceutical industry.