Pathway enrichment analysis approach based on topological structure and updated annotation of pathway.

Pathway enrichment analysis has been widely used to identify cancer risk pathways, and contributes to elucidating the mechanism of tumorigenesis. However, most of the existing approaches use the outdated pathway information and neglect the complex gene interactions in pathway. Here, we first reviewed the existing widely used pathway enrichment analysis approaches briefly, and then, we proposed a novel topology-based pathway enrichment analysis (TPEA) method, which integrated topological properties and global upstream/downstream positions of genes in pathways. We compared TPEA with four widely used pathway enrichment analysis tools, including database for annotation, visualization and integrated discovery (DAVID), gene set enrichment analysis (GSEA), centrality-based pathway enrichment (CePa) and signaling pathway impact analysis (SPIA), through analyzing six gene expression profiles of three tumor types (colorectal cancer, thyroid cancer and endometrial cancer). As a result, we identified several well-known cancer risk pathways that could not be obtained by the existing tools, and the results of TPEA were more stable than that of the other tools in analyzing different data sets of the same cancer. Ultimately, we developed an R package to implement TPEA, which could online update KEGG pathway information and is available at the Comprehensive R Archive Network (CRAN): https://cran.r-project.org/web/packages/TPEA/.

Accurate prediction and elucidation of drug resistance based on the robust and reproducible chemoresponse communities.

Selecting the available treatment for each cancer patient from genomic context is a core goal of precision medicine, but innovative approaches with mechanism interpretation and improved performance are still highly needed. Through utilizing in vitro chemotherapy response data coupled with gene and miRNA expression profiles, we applied a network-based approach that identified markers not as individual molecules but as functional groups extracted from the integrated transcription factor and miRNA regulatory network. Based on the identified chemoresponse communities, the predictors of drug resistance achieved high accuracy in cross-validation and were more robust and reproducible than conventional single-molecule markers. Meanwhile, as candidate communities not only enriched abundant cellular process but also covered a variety of drug enzymes, transporters, and targets, these resulting chemoresponse communities furnished novel models to interpret multiple kinds of potential regulatory mechanism, such as dysregulation of cancer cell apoptosis or disturbance of drug metabolism. Moreover, compounds were linked based on the enrichment of their common chemoresponse communities to uncover undetected response patterns and possible multidrug resistance phenotype. Finally, an omnibus repository named ChemoCommunity (http://www.jianglab.cn/ChemoCommunity/) was constructed, which furnished a user-friendly interface for a convenient retrieval of the detailed information on chemoresponse communities. Taken together, our method, and the accompanying database, improved the performance of classifiers for drug resistance and provided novel model to uncover the possible regulatory mechanism of individual response to drug.

Quantitative proteomics reveals the novel co-expression signatures in early brain development for prognosis of glioblastoma multiforme.

Although several researches have explored the similarity across development and tumorigenesis in cellular behavior and underlying molecular mechanisms, not many have investigated the developmental characteristics at proteomic level and further extended to cancer clinical outcome. In this study, we used iTRAQ to quantify the protein expression changes during macaque rhesus brain development from fetuses at gestation 70 days to after born 5 years. Then, we performed weighted gene co-expression network analysis (WGCNA) on protein expression data of brain development to identify co-expressed modules that highly expressed on distinct development stages, including early stage, middle stage and late stage. Moreover, we used the univariate cox regression model to evaluate the prognostic potentials of these genes in two independent glioblastoma multiforme (GBM) datasets. The results showed that the modules highly expressed on early stage contained more reproducible prognostic genes, including ILF2, CCT7, CCT4, RPL10A, MSN, PRPS1, TFRC and APEX1. These genes were not only associated with clinical outcome, but also tended to influence chemoresponse. These signatures identified from embryonic brain development might contribute to precise prediction of GBM prognosis and identification of novel drug targets in GBM therapies. Thus, the development could become a viable reference model for researching cancers, including identifying novel prognostic markers and promoting new therapies.

Dissecting the Origin of Breast Cancer Subtype Stem Cell and the Potential Mechanism of Malignant Transformation.

BACKGROUND: Breast cancer is the most common incident form of cancer in women including different subtypes. Cancer stem cells (CSCs) have been confirmed to exist in breast cancer. But the research on the origin of breast cancer subtype stem cells (BCSSCs) is still inadequate. METHODS: We identified the putative origin cells of BCSSCs through comparing gene signatures between BCSSCs and normal mammary cells from multiple perspectives: common signature, expression consistency, functional similarity and shortest path length. First, the potential origin cells were ranked according to these measures separately. Then Q statistic was employed to combine all rank lists into a unique list for each subtype, to prioritize the origin cells for each BCSSC. Next, we identified origin-related gene modules through integrating functional interaction network with differentially expressed genes. Finally, transcription factors of significant gene modules were predicted by MatchTM. RESULTS: The results showed that Luminal A CSC was most relevant to luminal progenitor cell or mature luminal cell; luminal B and HER2 CSC were most relevant to bipotent-enriched progenitor cell; basal-like CSC was most relevant to bipotent-enriched progenitor cell or mature luminal cell. Network modules analysis revealed genes related to mitochondrial respiratory chain (MRC) were significantly dysregulated during the origin of luminal B CSC. In addition, SOX10 emerged as a key regulator of MRC. CONCLUSIONS: Our study supports substantive evidence for the possible origin of four kinds of BCSSCs. Dysfunction of MRC may contribute to the origin of luminal B CSC. These findings may have important implications to treat and prevent breast cancer.

Psmir: a database of potential associations between small molecules and miRNAs.

miRNAs are key post-transcriptional regulators of many essential biological processes, and their dysregulation has been validated in almost all human cancers. Restoring aberrantly expressed miRNAs might be a novel therapeutics. Recently, many studies have demonstrated that small molecular compounds can affect miRNA expression. Thus, prediction of associations between small molecules and miRNAs is important for investigation of miRNA-targeted drugs. Here, we analyzed 39 miRNA-perturbed gene expression profiles, and then calculated the similarity of transcription responses between miRNA perturbation and drug treatment to predict drug-miRNA associations. At the significance level of 0.05, we obtained 6501 candidate associations between 1295 small molecules and 25 miRNAs, which included 624 FDA approved drugs. Finally, we constructed the Psmir database to store all potential associations and the related materials. In a word, Psmir served as a valuable resource for dissecting the biological significance in small molecules' effects on miRNA expression, which will facilitate developing novel potential therapeutic targets or treatments for human cancers. Psmir is supported by all major browsers, and is freely available at http://www.bio-bigdata.com/Psmir/.

The gain and loss of long noncoding RNA associated-competing endogenous RNAs in prostate cancer.

Prostate cancer (PC) is one of the most common solid tumors in men. However, the molecular mechanism of PC remains unclear. Numerous studies have demonstrated that long noncoding RNA (lncRNA) can act as microRNA (miRNA) sponge, one type of competing endogenous RNAs (ceRNAs), which offers a novel viewpoint to elucidate the mechanisms of PC. Here, we proposed an integrative systems biology approach to infer the gain and loss of ceRNAs in PC. First, we re-annotated exon microarray data to obtain lncRNA expression profiles of PC. Second, by integrating mRNA and miRNA expression, as well as miRNA targets, we constructed lncRNA-miRNA-mRNA ceRNA networks in cancer and normal samples. The lncRNAs in these two ceRNA networks tended to have a longer transcript length and cover more exons than the lncRNAs not involved in ceRNA networks. Next, we further extracted the gain and loss ceRNA networks in PC. We found that the gain ceRNAs in PC participated in cell cycle, and the loss ceRNAs in PC were associated with metabolism. We also identified potential prognostic ceRNA pairs such as MALAT1-EGR2 and MEG3-AQP3. Finally, we inferred a novel mechanism of known drugs, such as cisplatin, for the treatment of PC through gain and loss ceRNA networks. The potential drugs such as 1,2,6-tri-O-galloyl-beta-D-glucopyranose (TGGP) could modulate lncRNA-mRNA competing relationships, which may uncover new strategy for treating PC. In summary, we systematically investigated the gain and loss of ceRNAs in PC, which may prove useful for identifying potential biomarkers and therapeutics for PC.

Integrated systems approach identifies risk regulatory pathways and key regulators in coronary artery disease.

UNLABELLED: Coronary artery disease (CAD) is the most common type of heart disease. However, the molecular mechanisms of CAD remain elusive. Regulatory pathways are known to play crucial roles in many pathogenic processes. Thus, inferring risk regulatory pathways is an important step toward elucidating the mechanisms underlying CAD. With advances in high-throughput data, we developed an integrated systems approach to identify CAD risk regulatory pathways and key regulators. Firstly, a CAD-related core subnetwork was identified from a curated transcription factor (TF) and microRNA (miRNA) regulatory network based on a random walk algorithm. Secondly, candidate risk regulatory pathways were extracted from the subnetwork by applying a breadth-first search (BFS) algorithm. Then, risk regulatory pathways were prioritized based on multiple CAD-associated data sources. Finally, we also proposed a new measure to prioritize upstream regulators. We inferred that phosphatase and tensin homolog (PTEN) may be a key regulator in the dysregulation of risk regulatory pathways. This study takes a closer step than the identification of disease subnetworks or modules. From the risk regulatory pathways, we could understand the flow of regulatory information in the initiation and progression of the disease. Our approach helps to uncover its potential etiology. KEY MESSAGES: We developed an integrated systems approach to identify risk regulatory pathways. We proposed a new measure to prioritize the key regulators in CAD. PTEN may be a key regulator in dysregulation of the risk regulatory pathways.

EpimiR: a database of curated mutual regulation between miRNAs and epigenetic modifications.

As two kinds of important gene expression regulators, both epigenetic modification and microRNA (miRNA) can play significant roles in a wide range of human diseases. Recently, many studies have demonstrated that epigenetics and miRNA can affect each other in various ways. In this study, we established the EpimiR database, which collects 1974 regulations between 19 kinds of epigenetic modifications (such as DNA methylation, histone acetylation, H3K4me3, H3S10p) and 617 miRNAs across seven species (including Homo sapiens, Mus musculus, Rattus norvegicus, Gallus gallus, Epstein-Barr virus, Canis familiaris and Arabidopsis thaliana) from >300 references in the literature. These regulations can be divided into two parts: miR2Epi (103 entries describing how miRNA regulates epigenetic modification) and Epi2miR (1871 entries describing how epigenetic modification affects miRNA). Each entry of EpimiR not only contains basic descriptions of the validated experiment (method, species, reference and so on) but also clearly illuminates the regulatory pathway between epigenetics and miRNA. As a supplement to the curated information, the EpimiR extends to gather predicted epigenetic features (such as predicted transcription start site, upstream CpG island) associated with miRNA for users to guide their future biological experiments. Finally, EpimiR offers download and submission pages. Thus, EpimiR provides a fairly comprehensive repository about the mutual regulation between epigenetic modifications and miRNAs, which will promote the research on the regulatory mechanism of epigenetics and miRNA. Database URL: http://bioinfo.hrbmu.edu.cn/EpimiR/.

The pan-cancer analysis of gene expression patterns in the context of inflammation.

Although several studies have investigated the essential roles of inflammation in tumor progression, not many have systematically analyzed gene expression patterns across diverse cancers in the context of inflammation. In this study, in order to better understand the inflammatory scenario, we initially constructed the inflammatory timeline (IT) based on two gene expression profiles during inflammatory progression (inflammatory bowel disease and Helicobacter pylori infection). Then, we separately identified the differentially expressed genes (DEGs) from 25 cancer-related microarray data. By comparing the distributions of DEGs in the IT, we identified three novel pan-cancer gene expression patterns. In the first pattern, the up-regulated genes in cancers were over-expressed in the early phase of inflammation, while the down-regulated genes were over-expressed in the late phase of inflammation. The second pattern was the opposite of the first one. The third pattern appeared to be transitional between the first and second patterns. We found that some cancers with different tissue origins have similar gene expression patterns. Finally, we identified two sets of tissue-independent inflammatory signatures that were over-expressed in early and late phases of inflammation, respectively. The dominant biological processes of early inflammatory signatures were cell proliferation, DNA replication, and DNA repair, whereas the late inflammatory signatures were reflective of innate immune response, neutrophil migration, and antigen processing. These inflammatory signatures may be useful to predict gene expression patterns in human cancers. Therefore, the pan-cancer analysis of gene expression patterns in the context of inflammation provides a novel insight into cancers and an unprecedented opportunity to develop new therapies.