Search

Background Genome-wide association studies (GWASes) have identified many noncoding germline single-nucleotide polymorphisms (SNPs) that are associated with an increased risk of developing cancer. However, how these SNPs affect cancer risk is still largely unknown. Methods We used a systems biology approach to analyse the regulatory role of cancer-risk SNPs in thirteen tissues. By using data from the Genotype-Tissue Expression (GTEx) project, we performed an expression quantitative trait locus (eQTL) analysis. We represented both significant cis- and trans-eQTLs as edges in tissue-specific eQTL bipartite networks. Results Each tissue-specific eQTL network is organised into communities that group sets of SNPs and functionally related genes. When mapping cancer-risk SNPs to these networks, we find that in each tissue, these SNPs are significantly overrepresented in communities enriched for immune response processes, as well as tissue-specific functions. Moreover, cancer-risk SNPs are more likely to be ‘cores’ of their communities, influencing the expression of many genes within the same biological processes. Finally, cancer-risk SNPs preferentially target oncogenes and tumour-suppressor genes, suggesting that they may alter the expression of these key cancer genes. Conclusions This approach provides a new way of understanding genetic effects on cancer risk and provides a biological context for interpreting the results of GWAS cancer studies.

Sex differences manifest in many diseases and may drive sex-specific therapeutic responses. To understand the molecular basis of sex differences, we evaluated sex-biased gene regulation by constructing sample-specific gene regulatory networks in 29 human healthy tissues using 8,279 whole-genome expression profiles from the Genotype-Tissue Expression (GTEx) project. We find sex-biased regulatory network structures in each tissue. Even though most transcription factors (TFs) are not differentially expressed between males and females, many have sex-biased regulatory targeting patterns. In each tissue, genes that are differentially targeted by TFs between the sexes are enriched for tissue-related functions and diseases. In brain tissue, for example, genes associated with Parkinson’s disease and Alzheimer’s disease are targeted by different sets of TFs in each sex. Our systems-based analysis identifies a repertoire of TFs that play important roles in sex-specific architecture of gene regulatory networks, and it underlines sex-specific regulatory processes in both health and disease.

One of the biggest challenges in cancer is predicting its initiation and course of progression. In this issue of Cancer Research, Rockne and colleagues use state transition theory to predict how peripheral mononuclear blood cells in mice transition from a healthy state to acute myeloid leukemia. They found that critical transcriptomic perturbations could predict initiation and progression of the disease. This is an important step toward accurately predicting cancer evolution, which may eventually facilitate early diagnosis of cancer and disease recurrence, and which could potentially inform on timing of therapeutic interventions. See related article by Rockne et al., 3157

Motivation Conventional methods to analyze genomic data do not make use of the interplay between multiple factors, such as between microRNAs (miRNAs) and the messenger RNA (mRNA) transcripts they regulate, and thereby often fail to identify the cellular processes that are unique to specific tissues. We developed PUMA (PANDA Using MicroRNA Associations), a computational tool that uses message passing to integrate a prior network of miRNA target predictions with target gene co-expression information to model genome-wide gene regulation by miRNAs. We applied PUMA to 38 tissues from the Genotype-Tissue Expression project, integrating RNA-Seq data with two different miRNA target predictions priors, built on predictions from TargetScan and miRanda, respectively. We found that while target predictions obtained from these two different resources are considerably different, PUMA captures similar tissue-specific miRNA–target regulatory interactions in the different network models. Furthermore, the tissue-specific functions of miRNAs we identified based on regulatory profiles (available at: https://kuijjer.shinyapps.io/puma_gtex/) are highly similar between networks modeled on the two target prediction resources. This indicates that PUMA consistently captures important tissue-specific miRNA regulatory processes. In addition, using PUMA we identified miRNAs regulating important tissue-specific processes that, when mutated, may result in disease development in the same tissue. Availability and implementation PUMA is available in C++, MATLAB and Python on GitHub (https://github.com/kuijjerlab and https://netzoo.github.io/). Supplementary information Supplementary data are available at Bioinformatics online.

Based on morphology it is often challenging to distinguish between the many different soft tissue sarcoma subtypes. Moreover, outcome of disease is highly variable even between patients with the same disease. Machine learning on transcriptome sequencing data could be a valuable new tool to understand differences between and within entities. Here we used machine learning analysis to identify novel diagnostic and prognostic markers and therapeutic targets for soft tissue sarcomas. Gene expression data was used from the Cancer Genome Atlas, the Genotype-Tissue Expression project and the French Sarcoma Group. We identified three groups of tumors that overlap in their molecular profiles as seen with unsupervised t-Distributed Stochastic Neighbor Embedding clustering and a deep neural network. The three groups corresponded to subtypes that are morphologically overlapping. Using a random forest algorithm, we identified novel diagnostic markers for soft tissue sarcoma that distinguished between synovial sarcoma and MPNST, and that we validated using qRT-PCR in an independent series. Next, we identified prognostic genes that are strong predictors of disease outcome when used in a k-nearest neighbor algorithm. The prognostic genes were further validated in expression data from the French Sarcoma Group. One of these, HMMR, was validated in an independent series of leiomyosarcomas using immunohistochemistry on tissue micro array as a prognostic gene for disease-free interval. Furthermore, reconstruction of regulatory networks combined with data from the Connectivity Map showed, amongst others, that HDAC inhibitors could be a potential effective therapy for multiple soft tissue sarcoma subtypes. A viability assay with two HDAC inhibitors confirmed that both leiomyosarcoma and synovial sarcoma are sensitive to HDAC inhibition. In this study we identified novel diagnostic markers, prognostic markers and therapeutic leads from multiple soft tissue sarcoma gene expression datasets. Thus, machine learning algorithms are powerful new tools to improve our understanding of rare tumor entities., Author summary Soft-tissue sarcomas are a group of rare cancers that can be challenging to diagnose and treat. The morphology of the different soft-tissue sarcoma subtypes can overlap and the prognosis differs significantly between, and also within, the different subtypes. Moreover, targeted therapies are often not available. In this study we used transcriptome sequencing data from The Cancer Genome Atlas, containing 206 soft-tissue sarcoma samples which we analyzed using different machine learning algorithms to gain novel insights. When possible, we verified our findings in independent datasets or in cell lines. First, we found that both synovial sarcomas and malignant peripheral nerve sheath tumors show the largest overlap with normal tissue derived from the nervous system. This link with neural differentiation for synovial sarcoma was not well established until now. Second, genes were identified whose expression could be used to differentiate between the different soft-tissue sarcomas where the morphology overlaps. Third, novel prognostic genes were identified for separate subtypes. One gene, HMMR, which we found as a strong prognostic gene for leiomyosarcoma, was verified with immunohistochemistry on samples from our archives. Last, using a network analysis new potential therapies were identified. HDAC inhibitors were identified as a potential strong therapy for sarcomas, including leiomyosarcomas, which we verified in cell lines.

Background In biomedical research, network inference algorithms are typically used to infer complex association patterns between biological entities, such as between genes or proteins, using data from a population. This resulting aggregate network, in essence, averages over the networks of those individuals in the population. LIONESS (Linear Interpolation to Obtain Network Estimates for Single Samples) is a method that can be used together with a network inference algorithm to extract networks for individual samples in a population. The method’s key characteristic is that, by modeling networks for individual samples in a data set, it can capture network heterogeneity in a population. LIONESS was originally made available as a function within the PANDA (Passing Attributes between Networks for Data Assimilation) regulatory network reconstruction framework. However, the LIONESS algorithm is generalizable and can be used to model single sample networks based on a wide range of network inference algorithms. Results In this software article, we describe lionessR, an R implementation of LIONESS that can be applied to any network inference method in R that outputs a complete, weighted adjacency matrix. As an example, we provide a vignette of an application of lionessR to model single sample networks based on correlated gene expression in a bone cancer dataset. We show how the tool can be used to identify differential patterns of correlation between two groups of patients. Conclusions We developed lionessR, an open source R package to model single sample networks. We show how lionessR can be used to inform us on potential precision medicine applications in cancer. The lionessR package is a user-friendly tool to perform such analyses. The package, which includes a vignette describing the application, is freely available at: https://github.com/kuijjerlab/lionessR and at: http://bioconductor.org/packages/lionessR.

Genome-wide associations studies (GWASes) have identified many germline genetic variants that are associated with an increased risk of developing cancer. However, how these single nucleotide polymorphisms (SNPs) alter biological function in a way that increases cancer risk is still largely unknown. We used a systems biology approach to analyze the regulatory role and functional associations of cancer-risk SNPs in thirteen distinct tissues. Using data from the Genotype-Tissue Expression (GTEx) project, we performed an expression quantitative trait locus (eQTL) analysis, keeping both cis- and trans-eQTLs, and representing those significant associations as edges in tissue-specific eQTL bipartite networks. We find that each network is organized into highly modular communities that group sets of SNPs together with functionally-related collections of genes. We mapped cancer-risk SNPs to each tissue-specific eQTL network. Although we find in each tissue that cancer-risk SNPs are distributed across the network, they are not uniformly distributed. Rather they are significantly over-represented in a small number of communities. This includes communities enriched for immune response processes as well as communities representing tissue-specific functions. Moreover, cancer-risk SNPs are over-represented in the central “cores” of communities, meaning they are more likely to influence the expression of many genes within the same community, thus affecting biological processes. And finally, we find that cancer-risk SNPs preferentially target oncogenes and tumor suppressor genes, suggesting non-genic mutations may still alter the effects of these key cancer-associated genes. This bipartite eQTL network approach provides a new way of understanding genetic effects on cancer risk and provides a biological context for interpreting the results of GWAS cancer studies.

Significant sex differences are observed in colon cancer, and understanding these differences is essential to advance disease prevention, diagnosis, and treatment. Males have a higher risk of developing colon cancer and a lower survival rate than women. However, the molecular features that drive these sex differences are poorly understood. We used both transcript-based and gene regulatory network methods to analyze RNA-Seq data from The Cancer Genome Atlas for 445 patients with colon cancer. We compared gene expression between tumors in men and women and found no significant sex differences except for sex-chromosome genes. We then inferred patient-specific gene regulatory networks, and found significant regulatory differences between males and females, with drug and xenobiotics metabolism via cytochrome P450 pathways more strongly targeted in females. This finding was validated in a dataset that included 1,193 patients from five independent studies. While targeting of the drug metabolism pathway did not change the overall survival for males treated with adjuvant chemotherapy, females with greater targeting had an increase in 10-year overall survival probability, with 89% (95% CI: 78%-100%) survival compared to 61% (95% CI: 45%-82%) for women with lower targeting, respectively (p=0.034). Our network analysis uncovered patterns of transcriptional regulation that differentiate male and female colon cancer. Most importantly, targeting of the drug metabolism pathway was predictive of survival in women who received adjuvant chemotherapy. This network-based approach can be used to investigate the molecular features that drive sex differences in other cancers and complex diseases.

Characterizing the collective regulatory impact of genetic variants on complex phenotypes is a major challenge in developing a genotype to phenotype map. Using expression quantitative trait locus (eQTL) analyses, we constructed bipartite networks in which edges represent significant associations between genetic variants and gene expression levels and found that the network structure informs regulatory function. We show, in 13 tissues, that these eQTL networks are organized into dense, highly modular communities grouping genes often involved in coherent biological processes. We find communities representing shared processes across tissues, as well as communities associated with tissue-specific processes that coalesce around variants in tissue-specific active chromatin regions. Node centrality is also highly informative, with the global and community hubs differing in regulatory potential and likelihood of being disease associated.

Summary Although all human tissues carry out common processes, tissues are distinguished by gene expression patterns, implying that distinct regulatory programs control tissue specificity. In this study, we investigate gene expression and regulation across 38 tissues profiled in the Genotype-Tissue Expression project. We find that network edges (transcription factor to target gene connections) have higher tissue specificity than network nodes (genes) and that regulating nodes (transcription factors) are less likely to be expressed in a tissue-specific manner as compared to their targets (genes). Gene set enrichment analysis of network targeting also indicates that the regulation of tissue-specific function is largely independent of transcription factor expression. In addition, tissue-specific genes are not highly targeted in their corresponding tissue network. However, they do assume bottleneck positions due to variability in transcription factor targeting and the influence of non-canonical regulatory interactions. These results suggest that tissue specificity is driven by context-dependent regulatory paths, providing transcriptional control of tissue-specific processes., Graphical abstract

In the original version of [this article][1] ([1][2]), in the Disclosure of Potential Conflicts of Interest section, Dr. Charles S. Fuchs did not report consulting work with the biotechnology and pharmaceutical industry focused on cancer drug development. With respect to an analysis of molecular

SummaryExpression quantitative trait locus (eQTL) analysis associates genotype with gene expression, but most eQTL studies only includecis-acting variants and generally examine a single tissue. We used data from 13 tissues obtained by the Genotype-Tissue Expression (GTEx) project v6.0 and, in each tissue, identified bothcis- andtrans-eQTLs. For each tissue, we represented significant associations between single nucleotide polymorphisms (SNPs) and genes as edges in a bipartite network. These networks are organized into dense, highly modular communities often representing coherent biological processes. Global network hubs are enriched in distal gene regulatory regions such as enhancers, but are devoid of disease-associated SNPs from genome wide association studies. In contrast, local, community-specific network hubs (core SNPs) are preferentially located in regulatory regions such as promoters and enhancers and highly enriched for trait and disease associations. These results provide help explain how many weak-effect SNPs might together influence cellular function and phenotype.

PANDA (Passing Attributes between Networks for Data Assimilation) is a gene regulatory network inference method that uses message-passing to integrate multiple sources of 'omics data. PANDA was originally coded in C++. In this application note we describe PyPanda, the Python version of PANDA. PyPanda runs considerably faster than the C++ version and includes additional features for network analysis. Availability: The open source PyPanda Python package is freely available at https://github.com/davidvi/pypanda. Contact: d.g.p.van ijzendoorn@lumc.nl, 3 pages, 1 figure; version after Bioinformatics proofs

SummaryCell lines are an indispensable tool in biomedical research and often used as surrogates for tissues. An important question is how well a cell line’s transcriptional and regulatory processes reflect those of its tissue of origin. We analyzed RNA-Seq data from GTEx for 127 paired Epstein-Barr virus transformed lymphoblastoid cell lines and whole blood samples; and 244 paired fibroblast cell lines and skin biopsies. A combination of gene expression and network analyses shows that while cell lines carry the expression signatures of their primary tissues, albeit at reduced levels, they also exhibit changes in their patterns of transcription factor regulation. Cell cycle genes are over-expressed in cell lines compared to primary tissue, and they have a reduction of repressive transcription factor targeting. Our results provide insight into the expression and regulatory alterations observed in cell lines and suggest that these changes should be considered when using cell lines as models.HighlightsCell lines differ from their source tissues in gene expression and regulationDistinct cell lines share altered patterns of cell cycle regulationCell cycle genes are less strongly targeted by repressive TFs in cell linesCell lines share expression with their source tissue, but at reduced levels

SummarySexual dimorphism manifests in many diseases and may drive sex-specific therapeutic responses. To understand the molecular basis of sexual dimorphism, we conducted a comprehensive assessment of gene expression and regulatory network modeling in 31 tissues using 8716 human transcriptomes from GTEx. We observed sexually dimorphic patterns of gene expression involving as many as 60% of autosomal genes, depending on the tissue. Interestingly, sex hormone receptors do not exhibit sexually dimorphic expression in most tissues; however, differential network targeting by hormone receptors and other transcription factors (TFs) captures their downstream sexually dimorphic gene expression. Furthermore, differential network wiring was found extensively in several tissues, particularly in brain, in which not all regions exhibit strong differential expression. This systems-based analysis provides a new perspective on the drivers of sexual dimorphism, one in which a repertoire of TFs plays important roles in sex-specific rewiring of gene regulatory networks.HighlightsSexual dimorphism manifests in both gene expression and gene regulatory networksSubstantial sexual dimorphism in regulatory networks was found in several tissuesMany differentially regulated genes are not differentially expressedSex hormone receptors do not exhibit sexually dimorphic expression in most tissues

Although ultrahigh-throughput RNA-Sequencing has become the dominant technology for genome-wide transcriptional profiling, the vast majority of RNA-Seq studies typically profile only tens of samples, and most analytical pipelines are optimized for these smaller studies. However, projects are generating ever-larger data sets comprising RNA-Seq data from hundreds or thousands of samples, often collected at multiple centers and from diverse tissues. These complex data sets present significant analytical challenges due to batch and tissue effects, but provide the opportunity to revisit the assumptions and methods that we use to preprocess, normalize, and filter RNA-Seq data – critical first steps for any subsequent analysis. We find analysis of large RNA-Seq data sets requires both careful quality control and that one account for sparsity due to the heterogeneity intrinsic in multi-group studies. An R package instantiating our method for large-scale RNA-Seq normalization and preprocessing, YARN, is available at bioconductor.org/packages/yarn.HighlightsOverview of assumptions used in preprocessing and normalizationPipeline for preprocessing, quality control, and normalization of large heterogeneous dataA Bioconductor package for the YARN pipeline and easy manipulation of count dataPreprocessed GTEx data set using the YARN pipeline available as a resource

Summary Biological systems are driven by intricate interactions among molecules. Many methods have been developed that draw on large numbers of expression samples to infer connections between genes (or their products). The result is an aggregate network representing a single estimate for the likelihood of each interaction, or “edge,” in the network. Although informative, aggregate models fail to capture population heterogeneity. Here we propose a method to reverse engineer sample-specific networks from aggregate networks. We demonstrate our approach in several contexts, including simulated, yeast microarray, and human lymphoblastoid cell line RNA sequencing data. We use these sample-specific networks to study changes in network topology across time and to characterize shifts in gene regulation that were not apparent in the expression data. We believe that generating sample-specific networks will greatly facilitate the application of network methods to large, complex, and heterogeneous multi-omic datasets, supporting the emerging field of precision network medicine., Graphical Abstract, Highlights • We developed LIONESS to extract single-sample networks from aggregate models • We tested LIONESS using in silico, yeast, and human expression data • LIONESS-estimated networks are reproducible, accurate, and biologically meaningful • Single-sample network analysis highlights important biological processes, Biological Sciences; Bioinformatics; Complex Systems

Background In osteosarcoma, the presence of tumor-infiltrating macrophages positively correlates with patient survival in contrast to the negative effect of tumor-associated macrophages in patients with other tumors. Liposome-encapsulated muramyl tripeptide (L-MTP-PE) has been introduced in the treatment of osteosarcoma patients, which may enhance the potential anti-tumor activity of macrophages. Direct anti-tumor activity of human macrophages against human osteosarcoma cells has not been described so far. Hence, we assessed osteosarcoma cell growth after co-culture with human macrophages. Methods Monocyte-derived M1-like and M2-like macrophages were polarized with LPS + IFN-γ, L-MTP-PE +/− IFN-γ or IL-10 and incubated with osteosarcoma cells. Two days later, viable tumor cell numbers were analyzed. Antibody-dependent effects were investigated using the therapeutic anti-EGFR antibody cetuximab. Results M1-like macrophages inhibited osteosarcoma cell growth when activated with LPS + IFN-γ. Likewise, stimulation of M1-like macrophages with liposomal muramyl tripeptide (L-MTP-PE) inhibited tumor growth, but only when combined with IFN-γ. Addition of the tumor-reactive anti-EGFR antibody cetuximab did not further improve the anti-tumor activity of activated M1-like macrophages. The inhibition was mediated by supernatants of activated M1-like macrophages, containing TNF-α and IL-1β. However, specific blockage of these cytokines, nitric oxide or reactive oxygen species did not inhibit the anti-tumor effect, suggesting the involvement of other soluble factors released upon macrophage activation. While LPS + IFN-γ–activated M2-like macrophages had low anti-tumor activity, IL-10–polarized M2-like macrophages were able to reduce osteosarcoma cell growth in the presence of the anti-EGFR cetuximab involving antibody-dependent tumor cell phagocytosis. Conclusion This study demonstrates that human macrophages can be induced to exert direct anti-tumor activity against osteosarcoma cells. Our observation that the induction of macrophage anti-tumor activity by L-MTP-PE required IFN-γ may be of relevance for the optimization of L-MTP-PE therapy in osteosarcoma patients.

Ollier disease and Maffucci syndrome are non-hereditary skeletal disorders characterized by multiple enchondromas (Ollier disease) combined with spindle cell hemangiomas (Maffucci syndrome). We report somatic heterozygous mutations in IDH1 (c.394C>T encoding an R132C substitution and c.395G>A encoding an R132H substitution) or IDH2 (c.516G>C encoding R172S) in 87% of enchondromas (benign cartilage tumors) and in 70% of spindle cell hemangiomas (benign vascular lesions). In total, 35 of 43 (81%) subjects with Ollier disease and 10 of 13 (77%) with Maffucci syndrome carried IDH1 (98%) or IDH2 (2%) mutations in their tumors. Fourteen of 16 subjects had identical mutations in separate lesions. Immunohistochemistry to detect mutant IDH1 R132H protein suggested intraneoplastic and somatic mosaicism. IDH1 mutations in cartilage tumors were associated with hypermethylation and downregulated expression of several genes. Mutations were also found in 40% of solitary central cartilaginous tumors and in four chondrosarcoma cell lines, which will enable functional studies to assess the role of IDH1 and IDH2 mutations in tumor formation.

Osteosarcoma is the most frequent bone cancer in children and young adults. The outcome of patients with advanced disease is dismal. Exploitation of tumor-immune cell interactions may provide novel therapeutic approaches. CD70-CD27 interactions are important for the regulation of adaptive immunity. CD70 expression has been reported in some solid cancers and implicated in tumor escape from immunosurveillance. In this study, expression of CD70 and CD27 was analyzed in osteosarcoma cell lines and tumor specimens. CD70 protein was expressed on most osteosarcoma cell lines (5/7) and patient-derived primary osteosarcoma cultures (4/6) as measured by flow cytometry. In contrast, CD70 was detected on few Ewing sarcoma cell lines (5/15) and was virtually absent from neuroblastoma (1/7) and rhabdomyosarcoma cell lines (0/5). CD70(+) primary cultures were derived from CD70(+) osteosarcoma lesions. CD70 expression in osteosarcoma cryosections was heterogeneous, restricted to tumor cells and not attributed to infiltrating CD3(+) T cells as assessed by immunohistochemistry/immunofluorescence. CD70 was detected in primary (1/5) but also recurrent (2/4) and metastatic (1/3) tumors. CD27, the receptor for CD70, was neither detected on tumor cells nor on T cells in CD70(+) or CD70(-) tumors, suggesting that CD70 on tumor cells is not involved in CD27-dependent tumor-immune cell interactions in osteosarcoma. CD70 gene expression in diagnostic biopsies of osteosarcoma patients did not correlate with the occurrence of metastasis and survival (n = 70). Our data illustrate that CD70 is expressed in a subset of osteosarcoma patients. In patients with CD70(+) tumors, CD70 may represent a novel candidate for antibody-based targeted immunotherapy.