Search

Abstract Background Much of the public access cancer microarray data is asymmetric, belonging to datasets containing no samples from normal tissue. Asymmetric data cannot be used in standard meta-analysis approaches (such as the inverse variance method) to obtain large sample sizes for statistical power enrichment. Noting that plenty of normal tissue microarray samples exist in studies not involving cancer, we investigated the viability and accuracy of an integrated microarray analysis approach based on significance analysis of microarrays (merged SAM) using a collection of data from separate diseased and normal samples. Results We focused on five solid cancer types (colon, kidney, liver, lung, and pancreas), where available microarray data allowed us to compare meta-analysis and integrated approaches. Our results from the merged SAM significantly overlapped gene lists from the validated inverse-variance method. Both meta-analysis and merged SAM approaches successfully captured the aberrances in the cell cycle that commonly occur in the different cancer types. However, the integrated SAM analysis replicated the known cancer literature (excluding microarray studies) with much more accuracy than the meta-analysis. Conclusion The merged SAM test is a powerful, robust approach for combining data from similar platforms and for analyzing asymmetric datasets, including those with only normal or only cancer samples that cannot be utilized by meta-analysis methods. The integrated SAM approach can also be used in comparing global gene expression between various subtypes of cancer arising from the same tissue.

Abstract Background Phosphorylation events direct the flow of signals and metabolites along cellular protein networks. Current annotations of kinase-substrate binding events are far from complete. In this study, we scanned the entire human protein sequences using the PROSITE domain annotation tool to identify patterns of domain composition in kinases and their substrates. We identified statistically enriched pairs of strings of domains (signature pairs) in kinase-substrate couples presented in the 2006 version of the PTM database. Results The signature pairs enriched in kinase - substrate binding interactions turned out to be highly specific to kinase subtypes. The resulting list of signature pairs predicted kinase-substrate interactions in validation dataset not used in learning with high statistical accuracy. Conclusions The method presented here produces predictions of protein phosphorylation events with high accuracy and mid-level coverage. Our method can be used in expanding the currently available drafts of cell signaling pathways and thus will be an important tool in the development of combination drug therapies targeting complex diseases.

Abstract Background The HIV viral genome mutates at a high rate and poses a significant long term health risk even in the presence of combination antiretroviral therapy. Current methods for predicting a patient's response to therapy rely on site-directed mutagenesis experiments and in vitro resistance assays. In this bioinformatics study we treat response to antiretroviral therapy as a two-body problem: response to therapy is considered to be a function of both the host and pathogen proteomes. We set out to identify potential responders based on the presence or absence of host protein and DNA motifs on the HIV proteome. Results An alignment of thousands of HIV-1 sequences attested to extensive variation in nucleotide sequence but also showed conservation of eukaryotic short linear motifs on the protein coding regions. The reduction in viral load of patients in the Stanford HIV Drug Resistance Database exhibited a bimodal distribution after 24 weeks of antiretroviral therapy, with 2,000 copies/ml cutoff. Similarly, patients allocated into responder/non-responder categories based on consistent viral load reduction during a 24 week period showed clear separation. In both cases of phenotype identification, a set of features composed of short linear motifs in the reverse transcriptase region of HIV sequence accurately predicted a patient's response to therapy. Motifs that overlap resistance sites were highly predictive of responder identification in single drug regimens but these features lost importance in defining responders in multi-drug therapies. Conclusion HIV sequence mutates in a way that preferentially preserves peptide sequence motifs that are also found in the human proteome. The presence and absence of such motifs at specific regions of the HIV sequence is highly predictive of response to therapy. Some of these predictive motifs overlap with known HIV-1 resistance sites. These motifs are well established in bioinformatics databases and hence do not require identification via in vitro mutation experiments.

Abstract Background Host protein-protein interaction networks are altered by invading virus proteins, which create new interactions, and modify or destroy others. The resulting network topology favors excessive amounts of virus production in a stressed host cell network. Short linear peptide motifs common to both virus and host provide the basis for host network modification. Methods We focused our host-pathogen study on the binding and competing interactions of HIV-1 and human proteins. We showed that peptide motifs conserved across 70% of HIV-1 subtype B and C samples occurred in similar positions on HIV-1 proteins, and we documented protein domains that interact with these conserved motifs. We predicted which human proteins may be targeted by HIV-1 by taking pairs of human proteins that may interact via a motif conserved in HIV-1 and the corresponding interacting protein domain. Results Our predictions were enriched with host proteins known to interact with HIV-1 proteins ENV, NEF, and TAT (p-value < 4.26E-21). Cellular pathways statistically enriched for our predictions include the T cell receptor signaling, natural killer cell mediated cytotoxicity, cell cycle, and apoptosis pathways. Gene Ontology molecular function level 5 categories enriched with both predicted and confirmed HIV-1 targeted proteins included categories associated with phosphorylation events and adenyl ribonucleotide binding. Conclusion A list of host proteins highly enriched with those targeted by HIV-1 proteins can be obtained by searching for host protein motifs along virus protein sequences. The resulting set of host proteins predicted to be targeted by virus proteins will become more accurate with better annotations of motifs and domains. Nevertheless, our study validates the role of linear binding motifs shared by virus and host proteins as an important part of the crosstalk between virus and host.

Abstract Background Gene expression is controlled over a wide range at the transcript level through complex interplay between DNA and regulatory proteins, resulting in profiles of gene expression that can be represented as normal, graded, and bimodal (switch-like) distributions. We have previously performed genome-scale identification and annotation of genes with switch-like expression at the transcript level in mouse, using large microarray datasets for healthy tissue, in order to study the cellular pathways and regulatory mechanisms involving this class of genes. We showed that a large population of bimodal mouse genes encoding for cell membrane and extracellular matrix proteins is involved in communication pathways. This study expands on previous results by annotating human bimodal genes, investigating their correspondence to bimodality in mouse orthologs and exploring possible regulatory mechanisms that contribute to bimodality in gene expression in human and mouse. Results Fourteen percent of the human genes on the HGU133A array (1847 out of 13076) were identified as bimodal or switch-like. More than 40% were found to have bimodal mouse orthologs. KEGG pathways enriched for bimodal genes included ECM-receptor interaction, focal adhesion, and tight junction, showing strong similarity to the results obtained in mouse. Tissue-specific modes of expression of bimodal genes among brain, heart, and skeletal muscle were common between human and mouse. Promoter analysis revealed a higher than average number of transcription start sites per gene within the set of bimodal genes. Moreover, the bimodal gene set had differentially methylated histones compared to the set of the remaining genes in the genome. Conclusion The fact that bimodal genes were enriched within the cell membrane and extracellular environment make these genes as candidates for biomarkers for tissue specificity. The commonality of the important roles bimodal genes play in tissue differentiation in both the human and mouse indicates the potential value of mouse data in providing context for human tissue studies. The regulation motifs enriched in the bimodal gene set (TATA boxes, alternative promoters, methlyation) have known associations with complex diseases, such as cancer, providing further potential for the use of bimodal genes in studying the molecular basis of disease.

Abstract Background Large-scale compilation of gene expression microarray datasets across diverse biological phenotypes provided a means of gathering a priori knowledge in the form of identification and annotation of bimodal genes in the human and mouse genomes. These switch-like genes consist of 15% of known human genes, and are enriched with genes coding for extracellular and membrane proteins. It is of interest to determine the prediction potential of bimodal genes for class discovery in large-scale datasets. Results Use of a model-based clustering algorithm accurately classified more than 400 microarray samples into 19 different tissue types on the basis of bimodal gene expression. Bimodal expression patterns were also highly effective in differentiating between infectious diseases in model-based clustering of microarray data. Supervised classification with feature selection restricted to switch-like genes also recognized tissue specific and infectious disease specific signatures in independent test datasets reserved for validation. Determination of "on" and "off" states of switch-like genes in various tissues and diseases allowed for the identification of activated/deactivated pathways. Activated switch-like genes in neural, skeletal muscle and cardiac muscle tissue tend to have tissue-specific roles. A majority of activated genes in infectious disease are involved in processes related to the immune response. Conclusion Switch-like bimodal gene sets capture genome-wide signatures from microarray data in health and infectious disease. A subset of bimodal genes coding for extracellular and membrane proteins are associated with tissue specificity, indicating a potential role for them as biomarkers provided that expression is altered in the onset of disease. Furthermore, we provide evidence that bimodal genes are involved in temporally and spatially active mechanisms including tissue-specific functions and response of the immune system to invading pathogens.

Abstract Background Recent studies have placed gene expression in the context of distribution profiles including housekeeping, graded, and bimodal (switch-like). Single-gene studies have shown bimodal expression results from healthy cell signaling and complex diseases such as cancer, however developing a comprehensive list of human bimodal genes has remained a major challenge due to inherent noise in human microarray data. This study presents a two-component mixture analysis of mouse gene expression data for genes on the Affymetrix MG-U74Av2 array for the detection and annotation of switch-like genes. Two-component normal mixtures were fit to the data to identify bimodal genes and their potential roles in cell signaling and disease progression. Results Seventeen percent of the genes on the MG-U74Av2 array (1519 out of 9091) were identified as bimodal or switch-like. KEGG pathways significantly enriched for bimodal genes included ECM-receptor interaction, cell communication, and focal adhesion. Similarly, the GO biological process "cell adhesion" and cellular component "extracellular matrix" were significantly enriched. Switch-like genes were found to be associated with such diseases as congestive heart failure, Alzheimer's disease, arteriosclerosis, breast neoplasms, hypertension, myocardial infarction, obesity, rheumatoid arthritis, and type I and type II diabetes. In diabetes alone, over two hundred bimodal genes were in a different mode of expression compared to normal tissue. Conclusion This research identified and annotated bimodal or switch-like genes in the mouse genome using a large collection of microarray data. Genes with bimodal expression were enriched within the cell membrane and extracellular environment. Hundreds of bimodal genes demonstrated alternate modes of expression in diabetic muscle, pancreas, liver, heart, and adipose tissue. Bimodal genes comprise a candidate set of biomarkers for a large number of disease states because their expressions are tightly regulated at the transcription level.

Abstract Background Independently derived expression profiles of the same biological condition often have few genes in common. In this study, we created populations of expression profiles from publicly available microarray datasets of cancer (breast, lymphoma and renal) samples linked to clinical information with an iterative machine learning algorithm. ROC curves were used to assess the prediction error of each profile for classification. We compared the prediction error of profiles correlated with molecular phenotype against profiles correlated with relapse-free status. Prediction error of profiles identified with supervised univariate feature selection algorithms were compared to profiles selected randomly from a) all genes on the microarray platform and b) a list of known disease-related genes (a priori selection). We also determined the relevance of expression profiles on test arrays from independent datasets, measured on either the same or different microarray platforms. Results Highly discriminative expression profiles were produced on both simulated gene expression data and expression data from breast cancer and lymphoma datasets on the basis of ER and BCL-6 expression, respectively. Use of relapse-free status to identify profiles for prognosis prediction resulted in poorly discriminative decision rules. Supervised feature selection resulted in more accurate classifications than random or a priori selection, however, the difference in prediction error decreased as the number of features increased. These results held when decision rules were applied across-datasets to samples profiled on the same microarray platform. Conclusion Our results show that many gene sets predict molecular phenotypes accurately. Given this, expression profiles identified using different training datasets should be expected to show little agreement. In addition, we demonstrate the difficulty in predicting relapse directly from microarray data using supervised machine learning approaches. These findings are relevant to the use of molecular profiling for the identification of candidate biomarker panels.

Abstract Background Cell lines are used in experimental investigation of cancer but their capacity to represent tumor cells has yet to be quantified. The aim of the study was to identify significant alterations in pathway usage in cell lines in comparison with normal and tumor tissue. Methods This study utilized a pathway-specific enrichment analysis of publicly accessible microarray data and quantified the gene expression differences between cell lines, tumor, and normal tissue cells for six different tissue types. KEGG pathways that are significantly different between cell lines and tumors, cell lines and normal tissues and tumor and normal tissue were identified through enrichment tests on gene lists obtained using Significance Analysis of Microarrays (SAM). Results Cellular pathways that were significantly upregulated in cell lines compared to tumor cells and normal cells of the same tissue type included ATP synthesis, cell communication, cell cycle, oxidative phosphorylation, purine, pyrimidine and pyruvate metabolism, and proteasome. Results on metabolic pathways suggested an increase in the velocity nucleotide metabolism and RNA production. Pathways that were downregulated in cell lines compared to tumor and normal tissue included cell communication, cell adhesion molecules (CAMs), and ECM-receptor interaction. Only a fraction of the significantly altered genes in tumor-to-normal comparison had similar expressions in cancer cell lines and tumor cells. These genes were tissue-specific and were distributed sparsely among multiple pathways. Conclusion Significantly altered genes in tumors compared to normal tissue were largely tissue specific. Among these genes downregulation was a major trend. In contrast, cell lines contained large sets of significantly upregulated genes that were common to multiple tissue types. Pathway upregulation in cell lines was most pronounced over metabolic pathways including cell nucleotide metabolism and oxidative phosphorylation. Signaling pathways involved in adhesion and communication of cultured cancer cells were downregulated. The three way pathways comparison presented in this study brings light into the differences in the use of cellular pathways by tumor cells and cancer cell lines.

Abstract Background Tumor classification is inexact and largely dependent on the qualitative pathological examination of the images of the tumor tissue slides. In this study, our aim was to develop an automated computational method to classify Hematoxylin and Eosin (H&E) stained tissue sections based on cancer tissue texture features. Methods Image processing of histology slide images was used to detect and identify adipose tissue, extracellular matrix, morphologically distinct cell nuclei types, and the tubular architecture. The texture parameters derived from image analysis were then applied to classify images in a supervised classification scheme using histologic grade of a testing set as guidance. Results The histologic grade assigned by pathologists to invasive breast carcinoma images strongly correlated with both the presence and extent of cell nuclei with dispersed chromatin and the architecture, specifically the extent of presence of tubular cross sections. The two parameters that differentiated tumor grade found in this study were (1) the number density of cell nuclei with dispersed chromatin and (2) the number density of tubular cross sections identified through image processing as white blobs that were surrounded by a continuous string of cell nuclei. Classification based on subdivisions of a whole slide image containing a high concentration of cancer cell nuclei consistently agreed with the grade classification of the entire slide. Conclusion The automated image analysis and classification presented in this study demonstrate the feasibility of developing clinically relevant classification of histology images based on micro- texture. This method provides pathologists an invaluable quantitative tool for evaluation of the components of the Nottingham system for breast tumor grading and avoid intra-observer variability thus increasing the consistency of the decision-making process.