Expression Microarray Experiments Research Articles

Malignant Pleural Mesothelioma Interactome with 364 Novel Protein-Protein Interactions.

Simple SummaryInternal organs like the heart and lungs, and body cavities like the thoracic and abdominal cavities, are covered by a thin, slippery layer called the mesothelium. Malignant pleural mesothelioma (MPM) is an aggressive cancer of the lining of the lung, where genetics and asbestos exposure play a role. It is not diagnosable until it becomes invasive, offering only a short survival time to the patient. To help understand the role of the genes that relate to this disease most of which are poorly understood, we constructed the ‘MPM interactome’, including in it the protein-protein interactions that we predicted computationally and those that are previously known in the literature. Five novel protein-protein interactions (PPIs) were tested and validated experimentally. 85.65% of the interactome is supported by genetic variant, transcriptomic, and proteomic evidence. Comparative transcriptome analysis revealed 5 repurposable drugs targeting the interactome proteins. We make the interactome available on a freely accessible web application, Wiki-MPM.Malignant pleural mesothelioma (MPM) is an aggressive cancer affecting the outer lining of the lung, with a median survival of less than one year. We constructed an ‘MPM interactome’ with over 300 computationally predicted protein-protein interactions (PPIs) and over 2400 known PPIs of 62 literature-curated genes whose activity affects MPM. Known PPIs of the 62 MPM associated genes were derived from Biological General Repository for Interaction Datasets (BioGRID) and Human Protein Reference Database (HPRD). Novel PPIs were predicted by applying the HiPPIP algorithm, which computes features of protein pairs such as cellular localization, molecular function, biological process membership, genomic location of the gene, and gene expression in microarray experiments, and classifies the pairwise features as interacting or non-interacting based on a random forest model. We validated five novel predicted PPIs experimentally. The interactome is significantly enriched with genes differentially ex-pressed in MPM tumors compared with normal pleura and with other thoracic tumors, genes whose high expression has been correlated with unfavorable prognosis in lung cancer, genes differentially expressed on crocidolite exposure, and exosome-derived proteins identified from malignant mesothelioma cell lines. 28 of the interactors of MPM proteins are targets of 147 U.S. Food and Drug Administration (FDA)-approved drugs. By comparing disease-associated versus drug-induced differential expression profiles, we identified five potentially repurposable drugs, namely cabazitaxel, primaquine, pyrimethamine, trimethoprim and gliclazide. Preclinical studies may be con-ducted in vitro to validate these computational results. Interactome analysis of disease-associated genes is a powerful approach with high translational impact. It shows how MPM-associated genes identified by various high throughput studies are functionally linked, leading to clinically translatable results such as repurposed drugs. The PPIs are made available on a webserver with interactive user interface, visualization and advanced search capabilities.

Abstract 3441: Genome-wide gene expression analysis in asbestos-related non-small cell lung cancer

Abstract Recently we examined genome-wide DNA methylation in lung cancer from patients occupationally exposed to asbestos (Kettunen et al 2017 Int J Cancer). Distinct methylation changes in lung tumors were shown in relation to asbestos exposure and tobacco smoking. Now we have performed genome-wide gene expression microarray experiments with Agilent SurePrint G3 Human GE 8x60K v3 Microarrays using same samples. Patients had been interviewed for work and smoking histories and their pulmonary asbestos fiber counts were measured by electron microscopy. Fresh-frozen lung cancer tissues (LC, n=42) and paired microscopically normal peripheral lung tissues (NT, n=21) were studied. After data processing and analyses for differential gene expression, data were filtered for adjusted p < 0.05 and fold change > 1.5. Principal component analysis showed distinct clusters of normal lung and cancer tissues. In lung cancers compared with their paired microscopically normal lung, 5931 genes such as those involved in neutrophil degranulation and immune system were differentially expressed, with the lowest adjusted p=6.71E-37. In our preliminary analyses, 34 genes in NT and 44 in LC, and 152 genes in NT and 12 in LC showed differential expression when the asbestos-exposed were compared to non-exposed and current smokers to ex-smokers, respectively. The genes differentially expressed in relation to smoking or to asbestos exposure had the lowest adjusted p=8.50E-06 and 9.83E-04, respectively. Within differential gene expression related to asbestos exposure, chemokine activity and regulation of cellular protein metabolic processes were implicated whereas microtubules and positive regulation of transcription were related with tobacco smoking burden. In conclusion, comparison of gene expression patterns between lung cancer tissue and paired normal lung showed distinct principal component analysis clusters. Our preliminary data suggest different major biological processes in gene expression for these exposures. Financial support from the Cancer Society of Finland (2014 and 2015) is gratefully acknowledged. Citation Format: Eeva Kettunen, Veer Singh Marwah, Henrik Wolff, Dario Greco, Kirsti Husgafvel-Pursiainen. Genome-wide gene expression analysis in asbestos-related non-small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3441.

Testing for Inequality Constraints in Singular Models by Trimming or Winsorizing the Variance Matrix

ABSTRACTThere are many applications in which a statistic follows, at least asymptotically, a normal distribution with a singular or nearly singular variance matrix. A classic example occurs in linear regression models under multicollinearity but there are many more such examples. There is well-developed theory for testing linear equality constraints when the alternative is two-sided and the variance matrix is either singular or nonsingular. In recent years, there is considerable, and growing, interest in developing methods for situations in which the estimated variance matrix is nearly singular. However, there is no corresponding methodology for addressing one-sided, that is, constrained or ordered alternatives. In this article, we develop a unified framework for analyzing such problems. Our approach may be viewed as the trimming or winsorizing of the eigenvalues of the corresponding variance matrix. The proposed methodology is applicable to a wide range of scientific problems and to a variety of statistical models in which inequality constraints arise. We illustrate the methodology using data from a gene expression microarray experiment obtained from the NIEHS’ Fibroid Growth Study. Supplementary materials for this article are available online.

Commonly used statistical methods for detecting differential gene expression in microarray experiments

Commonly used statistical methods for detecting differential gene expression in microarray experiments

ANP32A Regulates Histone 3 Acetylation and Promotes Leukemogenesis in AML

Leukemia initiation and progression not only depend on genetic alterations but also on abnormal epigenetic changes that cause disorder of gene expression. Acidic leucine-rich nuclear phosphoprotein-32A (ANP32A) binds directly to histone tails, preferentially to unmodified histone H3 tails (Schneider et al., 2004; Seo et al., 2002). Although ANP32A inhibits histone acetyltransferase activity in vitro (Seo et al., 2002), loss of ANP32A leads to decreased histone 3 acetylation (acetyl-H3) levels on interferon-stimulated genes in human cervical carcinoma HeLa S3 cells (Kadota et al., 2011). Aberrant ANP32A expression has been observed in many solid tumors, which involves in diverse processes depending on the cell context (Brody et al., 2004; Hoffarth et al., 2008), yet mechanisms are still unknown. Furthermore, there has as yet been no specific investigation into the role of ANP32A in acute myeloid leukemia (AML).We found that ANP32A expression was upregulated at both mRNA and protein levels in patients with AML. In AML primary cells from patients, ANP32A KD decreased cell proliferation and colony formation ability. In addition, ANP32A KD induced apoptosis and cell cycle arrest at G1 phase accompanied with the downregulated protein expression of survival or cell cycle progress genes BCL2, CDK4, CCND1 and the upregulated protein expression of pro-apoptosis genes BAK, BAD, cleaved caspase 3. These findings were also verified in various subtypes of AML cell lines and Mll-Af9 pre-leukemic cells. Thus, ANP32A is essential for cell growth via the promotion of cell-cycle progression and inhibition of apoptosis in AML.Given that ANP32A associates with histone acetylation modification, we analyzed the effect of ANP32A depletion on acetyl-H3 and histone 4 acetylation (acetyl-H4). Western blot analysis showed that ANP32A KD caused global decrease of acetyl-H3, but no effect on the global level of acetyl-H4. To investigate the mechanism of ANP32A in AML, ChIP-seq and gene expression microarray experiments were performed to determine acetyl-H3 enrichment and transcriptome profiles, respectively, in ANP32A KD and control K562 cells. We found a positive correlation between changes in gene expression and changes in acetyl-H3 (r=0.16, p<10-4), especially for differential expression genes (DEGs; r=0.40, p<10-4). Moreover, we also found genes with decreased acetyl-H3 enrichment displayed lower expression levels compared with genes with increased acetyl-H3 enrichment (p<10-3). Thus, ANP32A alters acetyl-H3 modification in association with changes in gene expression.Gene set enrichment analysis (GSEA) revealed that ANP32A KD significantly decreased the acetyl-H3 enrichment levels at genes of lipid metabolism signatures (e.g. lipid metabolic process, cholesterol homeostasis). Transcriptome profile analysis using GSEA showed that gene expression levels of lipid metabolic process signature were also reduced after ANP32A KD. We further identified 4 genes (APOC1, CNKSR3, H19, PCSK9) which had significantly lower levels of mRNA expression (p<10-2, FDR<10-2, FC>1.5) and decreased levels of acetyl-H3 enrichment (p<10-2, FDR<10-2, FC>1.5) in ANP32A KD cells. Apolipoprotein CI(APOC1) was the most significantly downregulated gene among these 4 genes, with decreased acetyl-H3 on its promoter. To explore whether lipid metabolism as a downstream effector of ANP32A contributes to ANP32A function in AML, re-introduction of APOC1 was performed in ANP32A KD THP1 or K562 cells. Re-introduction of APOC1 significantly rescued the inhibition of cell growth in ANP32A KD AML cells, also upregulated protein expression of BCL2, CDK4 and CCND1, showing that ANP32A functions in AML partially depending on APOC1. Importantly, APOC1 depletion decreased cell proliferation of AML cells in vitro, consistent with the point of view that complex lipid alterations contribute to endoplasmic reticulum stress-induced apoptosis in cancer cells (Beloribi-Djefaflia et al., 2016; Iurlaro et al., 2016).Overall, our data suggest ANP32A as a novel regulator of leukemogenesis in AML. ANP32A may involve in the regulation of genome-wide histone 3 acetylation that alters the expression of multiple genes, including lipid metabolism related gene APOC1. Thus, ANP32A is a promising drug target for AML therapy. DisclosuresNo relevant conflicts of interest to declare.

Genomic and transcriptomic profiles and in vitro resistance to mitoxantrone and idarubicin in pediatric acute leukemias.

A major problem in the treatment of leukemia is the development of drug resistance to chemotherapeutic agents. To determine the ex vivo drug resistance profile to anthracyclines, an 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazoliumbromide (MTT) cytotoxicity assay was performed on mononuclear cells obtained from 155 patients with acute lymphoblastic leukemia (ALL) or acute myeloblastic leukemia (AML). Gene expression profiles (for 51 patients with ALL and 16 with AML) were prepared on the basis of cRNA hybridization to oligonucleotide arrays of the human genome (Affymetrix). Hierarchical clustering, assignment location and biological function were investigated during the correlation analysis for identified probe sets. Comparative genomic hybridization (CGH) array profiles (34 patients with ALL and 12 with AML) were prepared on the basis of DNA hybridization to oligonucleotide arrays of the human genome (Agilent). The validation of the array results was performed by a quantitative reverse transcriptase polymerase chain reaction. The collected expression and CGH microarray experiment results indicate that the ITGB2, SCL6A7, CASP1 and DUSP genes may comprise a resistance marker for acute leukemia cells correlated with anthracyclines. Moreover, there were also identified chromosome rearrangements associated with drug resistance, such as del5q32-35.3 and amp8p12-p11.21. Precise genes, as well as genome aberrations, might be classified as targets in therapy. In AML, the resistance of blasts to idarubicin and mitoxantrone may reflect an impaired integrin pathway. In ALL, the development of resistance is caused by the inhibition of B and T cell activation. Copyright © 2016 John Wiley & Sons, Ltd.

Long Noncoding RNAs in HPV-Induced Oncogenesis

Long noncoding RNAs (lncRNAs) play important roles in a wide range of oncogenic processes, including malignant transformation, epigenetic reprogramming, epithelial-to-mesenchymal transition, and metastasis development. LncRNAs induced by oncogenic viral proteins were shown to play critical roles in tumor initiation and progression. Despite this, little is known about Human papillomavirus (HPV)-induced modulation of host’s lncRNAs. In this review, we gathered published information about altered lncRNAs upon HPV status (infection/protein activity), making use of descriptive research works and published gene expression microarray experiments. A diversity of lncRNAs demonstrated to be altered, including metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), H19, and maternally expressed gene 3 (MEG3). Their functions in several cancers were reviewed, indicating that they may represent potential candidates for future research on HPV-induced oncogenesis.

The Saccharomyces Genome Database: Advanced Searching Methods and Data Mining.

At the core of the Saccharomyces Genome Database (SGD) are chromosomal features that encode a product. These include protein-coding genes and major noncoding RNA genes, such as tRNA and rRNA genes. The basic entry point into SGD is a gene or open-reading frame name that leads directly to the locus summary information page. A keyword describing function, phenotype, selective condition, or text from abstracts will also provide a door into the SGD. A DNA or protein sequence can be used to identify a gene or a chromosomal region using BLAST. Protein and DNA sequence identifiers, PubMed and NCBI IDs, author names, and function terms are also valid entry points. The information in SGD has been gathered and is maintained by a group of scientific biocurators and software developers who are devoted to providing researchers with up-to-date information from the published literature, connections to all the major research resources, and tools that allow the data to be explored. All the collected information cannot be represented or summarized for every possible question; therefore, it is necessary to be able to search the structured data in the database. This protocol describes the YeastMine tool, which provides an advanced search capability via an interactive tool. The SGD also archives results from microarray expression experiments, and a strategy designed to explore these data using the SPELL (Serial Pattern of Expression Levels Locator) tool is provided.

A hybrid imputation approach for microarray missing value estimation

BackgroundMissing data is an inevitable phenomenon in gene expression microarray experiments due to instrument failure or human error. It has a negative impact on performance of downstream analysis. Technically, most existing approaches suffer from this prevalent problem. Imputation is one of the frequently used methods for processing missing data. Actually many developments have been achieved in the research on estimating missing values. The challenging task is how to improve imputation accuracy for data with a large missing rate.MethodsIn this paper, induced by the thought of collaborative training, we propose a novel hybrid imputation method, called Recursive Mutual Imputation (RMI). Specifically, RMI exploits global correlation information and local structure in the data, captured by two popular methods, Bayesian Principal Component Analysis (BPCA) and Local Least Squares (LLS), respectively. Mutual strategy is implemented by sharing the estimated data sequences at each recursive process. Meanwhile, we consider the imputation sequence based on the number of missing entries in the target gene. Furthermore, a weight based integrated method is utilized in the final assembling step.ResultsWe evaluate RMI with three state-of-art algorithms (BPCA, LLS, Iterated Local Least Squares imputation (ItrLLS)) on four publicly available microarray datasets. Experimental results clearly demonstrate that RMI significantly outperforms comparative methods in terms of Normalized Root Mean Square Error (NRMSE), especially for datasets with large missing rates and less complete genes.ConclusionsIt is noted that our proposed hybrid imputation approach incorporates both global and local information of microarray genes, which achieves lower NRMSE values against to any single approach only. Besides, this study highlights the need for considering the imputing sequence of missing entries for imputation methods.

Dietary nitrate is a modifier of vascular gene expression in old male mice.

Aging leads to a number of disadvantageous changes in the cardiovascular system. Deterioration of vascular homoeostasis with increase in oxidative stress, chronic low-grade inflammation, and impaired nitric oxide bioavailability results in endothelial dysfunction, increased vascular stiffness, and compromised arterial-ventricular interactions. A chronic dietary supplementation with the micronutrient nitrate has been demonstrated to improve vascular function. Healthy dietary patterns may regulate gene expression profiles. However, the mechanisms are incompletely understood. The changes that occur at the gene expression level and transcriptional profile following a nutritional modification with nitrate have not been elucidated. To determine the changes of the vascular transcriptome, we conducted gene expression microarray experiments on aortas of old mice, which were treated with dietary nitrate. Our results highlight differentially expressed genes overrepresented in gene ontology categories. Molecular interaction and reaction pathways involved in the calcium-signaling pathway and the detoxification system were identified. Our results provide novel insight to an altered gene-expression profile in old mice following nitrate supplementation. This supports the general notion of nutritional approaches to modulate age-related changes of vascular functions and its detrimental consequences.

New target genes in endometrial tumors show a role for the estrogen-receptor pathway in microsatellite-unstable cancers.

Microsatellite instability (MSI) in tumors results in an accumulation of mutations in (target) genes. Previous studies suggest that the profile of target genes differs according to tumor type. This paper describes the first genome-wide search for target genes for mismatch repair-deficient endometrial cancers. Genes expressed in normal endometrium containing coding repeats were analyzed for mutations in tumors. We identified 44 possible genes of which seven are highly mutated (>15%). Some candidates were also found mutated in colorectal and gastric tumors. The most frequently mutated gene, NRIP1 encoding nuclear receptor-interacting protein 1, was silenced in an endometrial tumor cell line and expression microarray experiments were performed. Silencing of NRIP1 was associated with differences in the expression of several genes in the estrogen-receptor network. Furthermore, an enrichment of genes related to cell cycle (regulation) and replication was observed. We present a new profile of target genes, some of them tissue specific, whereas others seem to play a more general role in MSI tumors. The high-mutation frequency combined with the expression data suggest, for the first time, an involvement of NRIP1 in endometrial cancer development.

The histone demethylase activity of Rph1 is not essential for its role in the transcriptional response to nutrient signaling.

Rph1 and Gis1 are two related yeast zinc finger proteins that function as downstream effectors in the Ras/PKA, TOR and Sch9 nutrient signaling pathways. Both proteins also contain JmjC histone demethylase domains, but only Rph1 is known to be an active enzyme, demethylating lysine 36 of histone H3. We have studied to what extent the demethylase activity of Rph1 contributes to its role in nutrient signaling by performing gene expression microarray experiments on a yeast strain containing a catalytically inactive allele of RPH1. We find that the enzymatic activity of Rph1 is not essential for its role in growth phase dependent gene regulation. However, the ability of Rph1 to both activate and repress transcription is partially impaired in the active site mutant, indicating that the demethylase activity may enhance its function in vivo. Consistent with this, we find that the Rph1 mutation and a deletion of the histone H3 methylase Set2 affect the same target genes in opposite directions. Genes that are differentially expressed in the Rph1 mutant are also enriched for binding of Rpd3, a downstream effector in silencing, to their promoters. The expression of some subtelomeric genes and genes involved in sporulation and meiosis are also affected by the mutation, suggesting a role for Rph1-dependent demethylation in regulating these genes. A small set of genes are more strongly affected by the active site mutation, indicating a more pronounced role for the demethylase activity in their regulation by Rph1.

Functional genomics annotation of a statistical epistasis network associated with bladder cancer susceptibility.

BackgroundSeveral different genetic and environmental factors have been identified as independent risk factors for bladder cancer in population-based studies. Recent studies have turned to understanding the role of gene-gene and gene-environment interactions in determining risk. We previously developed the bioinformatics framework of statistical epistasis networks (SEN) to characterize the global structure of interacting genetic factors associated with a particular disease or clinical outcome. By applying SEN to a population-based study of bladder cancer among Caucasians in New Hampshire, we were able to identify a set of connected genetic factors with strong and significant interaction effects on bladder cancer susceptibility.FindingsTo support our statistical findings using networks, in the present study, we performed pathway enrichment analyses on the set of genes identified using SEN, and found that they are associated with the carcinogen benzo[a]pyrene, a component of tobacco smoke. We further carried out an mRNA expression microarray experiment to validate statistical genetic interactions, and to determine if the set of genes identified in the SEN were differentially expressed in a normal bladder cell line and a bladder cancer cell line in the presence or absence of benzo[a]pyrene. Significant nonrandom sets of genes from the SEN were found to be differentially expressed in response to benzo[a]pyrene in both the normal bladder cells and the bladder cancer cells. In addition, the patterns of gene expression were significantly different between these two cell types.ConclusionsThe enrichment analyses and the gene expression microarray results support the idea that SEN analysis of bladder in population-based studies is able to identify biologically meaningful statistical patterns. These results bring us a step closer to a systems genetic approach to understanding cancer susceptibility that integrates population and laboratory-based studies.

Identification of Key Genes in the Response toSalmonella enterica Enteritidis,Salmonella enterica Pullorum, and Poly(I:C) in Chicken Spleen and Caecum

Salmonella enterica Enteritidis (S. Enteritidis) and Salmonella enterica Pullorum (S. pullorum) are regarded as a threat to poultry production. This study's aim is to characterize the expression profiles in response to three different challenges and to identify infection-related genes in the chicken spleen and caecum. Groups of the Chinese chicken breed Langshan were challenged with either S. Enteritidis, S. pullorum, or poly(I:C). The concentrations of cytokines and antibodies and the Salmonella colonization level of the caecum and liver were detected in each group at 7 days postinfection. Expression microarray experiments were conducted using mRNA isolated from both spleen and caecum. Crucial differentially expressed genes (DEGs) associated with immunity were identified. Four DEGs were identified in spleen of all three challenge groups (RBM16, FAH, SOX5, and RBM9) and different four genes in caecum (SOUL, FCN2, ANLN, and ACSL1). Expression profiles were clearly different among the three challenged groups. Genes enriched in the spleen of birds infected with S. pullorum were enriched in lymphocyte proliferation related pathways, but the enriched genes in the caecum of the same group were primarily enriched in innate immunity or antibacterial responses. The DEGs that appear across all three challenge groups might represent global response factors for different pathogens.

Thrombin selectively induces transcription of genes in human monocytes involved in inflammation and wound healing.

Thrombin is essential for blood coagulation but functions also as a mediator of cellular signalling. Gene expression microarray experiments in human monocytes revealed thrombin-induced upregulation of a limited subset of genes, which are almost exclusively involved in inflammation and wound healing. Among these, the expression of F3 gene encoding for tissue factor (TF) was enhanced indicating that this physiological initiator of coagulation cascade may create a feed-forward loop to enhance blood coagulation. Activation of protease-activated receptor type 1 (PAR1) was shown to play a main role in promoting TF expression. Moreover, thrombin induced phosphorylation of ERK1/2, an event that is required for expression of thrombin-regulated genes. Thrombin also increased the expression of TF at the protein level in monocytes as evidenced by Western blot and immunostaining. Furthermore, FXa generation induced by thrombin-stimulated monocytes was abolished by a TF blocking antibody and therefore it is entirely attributable to the expression of tissue factor. This cellular activity of thrombin provides a new molecular link between coagulation, inflammation and wound healing.

Age-related vascular gene expression profiling in mice

Increasing age involves a number of detrimental changes in the cardiovascular system and particularly on the large arteries. It deteriorates vascular integrity and leads to increased vascular stiffness entailing hypertension with increased cardiovascular morbidity and mortality. The consequences of continuous oxidative stress and damages to biomolecules include altered gene expression, genomic instability, mutations, loss of cell division and cellular responses to increased stress. Many studies have been performed in aged C57BL/6 mice; however, analyses of the age-related changes that occur at a gene expression level and transcriptional profile in vascular tissue have not been elucidated in depth. To determine the changes of the vascular transcriptome, we conducted gene expression microarray experiments on aortas of adult and old mice, in which age-related vascular dysfunction was confirmed by increased stiffness and associated systolic hypertension. Our results highlight differentially expressed genes overrepresented in Gene Ontology categories. Molecular interaction and reaction pathways involved in vascular functions and disease, within the transforming growth factor-beta (TGF-β) pathway, the renin-angiotensin system and the detoxification systems are displayed. Our results provide insight to an altered gene expression profile related to age, thus offering useful clues to counteract or prevent vascular aging and its detrimental consequences.

A Reverse Engineering Approach to Optimize Experiments for the Construction of Biological Regulatory Networks

One of the major objectives in systems biology is to understand the relation between the topological structures and the dynamics of biological regulatory networks. In this context, various mathematical tools have been developed to deduct structures of regulatory networks from microarray expression data. In general, from a single data set, one cannot deduct the whole network structure; additional expression data are usually needed. Thus how to design a microarray expression experiment in order to get the most information is a practical problem in systems biology. Here we propose three methods, namely, maximum distance method, trajectory entropy method, and sampling method, to derive the optimal initial conditions for experiments. The performance of these methods is tested and evaluated in three well-known regulatory networks (budding yeast cell cycle, fission yeast cell cycle, and E. coli. SOS network). Based on the evaluation, we propose an efficient strategy for the design of microarray expression experiments.

Research Resource: The Endometrium Database Resource (EDR)

In order to understand the biology of the endometrium and potentially develop new diagnostic tools and treatments for endometrial diseases, the highly orchestrated gene expression/regulation that occurs within the uterus must first be understood. Even though a wealth of information on endometrial gene expression/regulation is available, this information is scattered across several different resources in formats that can be difficult for the average bench scientist to query, integrate, and utilize. The Endometrium Database Resource (EDR) was created as a single evolving resource for protein- and micro-RNA-encoding genes that have been shown by gene expression microarray, Northern blot, or other experiments in the literature to have their expression regulated in the uterus of humans, mice, rats, cows, domestic pigs, guinea pigs, and sheep. Genes are annotated in EDR with basic gene information (eg, gene symbol and chromosome), gene orthologs, and gene ontologies. Links are also provided to external resources for publication/s, nucleic and amino acid sequence, gene product function, and Gene Expression Omnibus (GEO) phase expression graph information. The resource also allows for direct comparison of relative gene expression in different microarray experiments for genes shown in the literature to be differentially expressed in the uterus. It is available via a user-friendly, web-based interface and is available without charge or restriction to the entire scientific community. The EDR can be accessed at http://edr.research.bcm.edu.

First survey and functional annotation of prohormone and convertase genes in the pig

BackgroundThe pig is a biomedical model to study human and livestock traits. Many of these traits are controlled by neuropeptides that result from the cleavage of prohormones by prohormone convertases. Only 45 prohormones have been confirmed in the pig. Sequence homology can be ineffective to annotate prohormone genes in sequenced species like the pig due to the multifactorial nature of the prohormone processing. The goal of this study is to undertake the first complete survey of prohormone and prohormone convertases genes in the pig genome. These genes were functionally annotated based on 35 gene expression microarray experiments. The cleavage sites of prohormone sequences into potentially active neuropeptides were predicted.ResultsWe identified 95 unique prohormone genes, 2 alternative calcitonin-related sequences, 8 prohormone convertases and 1 cleavage facilitator in the pig genome 10.2 assembly and trace archives. Of these, 11 pig prohormone genes have not been reported in the UniProt, UniGene or Gene databases. These genes are intermedin, cortistatin, insulin-like 5, orexigenic neuropeptide QRFP, prokineticin 2, prolactin-releasing peptide, parathyroid hormone 2, urocortin, urocortin 2, urocortin 3, and urotensin 2-related peptide. In addition, a novel neuropeptide S was identified in the pig genome correcting the previously reported pig sequence that is identical to the rabbit sequence. Most differentially expressed prohormone genes were under-expressed in pigs experiencing immune challenge relative to the un-challenged controls, in non-pregnant relative to pregnant sows, in old relative to young embryos, and in non-neural relative to neural tissues. The cleavage prediction based on human sequences had the best performance with a correct classification rate of cleaved and non-cleaved sites of 92% suggesting that the processing of prohormones in pigs is similar to humans. The cleavage prediction models did not find conclusive evidence supporting the production of the bioactive neuropeptides urocortin 2, urocortin 3, torsin family 2 member A, tachykinin 4, islet amyloid polypeptide, and calcitonin receptor-stimulating peptide 2 in the pig.ConclusionsThe present genomic and functional characterization supports the use of the pig as an effective animal model to gain a deeper understanding of prohormones, prohormone convertases and neuropeptides in biomedical and agricultural research.

Distributional fold change test – a statistical approach for detecting differential expression in microarray experiments

BackgroundBecause of the large volume of data and the intrinsic variation of data intensity observed in microarray experiments, different statistical methods have been used to systematically extract biological information and to quantify the associated uncertainty. The simplest method to identify differentially expressed genes is to evaluate the ratio of average intensities in two different conditions and consider all genes that differ by more than an arbitrary cut-off value to be differentially expressed. This filtering approach is not a statistical test and there is no associated value that can indicate the level of confidence in the designation of genes as differentially expressed or not differentially expressed. At the same time the fold change by itself provide valuable information and it is important to find unambiguous ways of using this information in expression data treatment.ResultsA new method of finding differentially expressed genes, called distributional fold change (DFC) test is introduced. The method is based on an analysis of the intensity distribution of all microarray probe sets mapped to a three dimensional feature space composed of average expression level, average difference of gene expression and total variance. The proposed method allows one to rank each feature based on the signal-to-noise ratio and to ascertain for each feature the confidence level and power for being differentially expressed. The performance of the new method was evaluated using the total and partial area under receiver operating curves and tested on 11 data sets from Gene Omnibus Database with independently verified differentially expressed genes and compared with the t-test and shrinkage t-test. Overall the DFC test performed the best – on average it had higher sensitivity and partial AUC and its elevation was most prominent in the low range of differentially expressed features, typical for formalin-fixed paraffin-embedded sample sets.ConclusionsThe distributional fold change test is an effective method for finding and ranking differentially expressed probesets on microarrays. The application of this test is advantageous to data sets using formalin-fixed paraffin-embedded samples or other systems where degradation effects diminish the applicability of correlation adjusted methods to the whole feature set.