Search

Abstract Background Genome wide gene expression data is a rich source for the identification of gene signatures suitable for clinical purposes and a number of statistical algorithms have been described for both identification and evaluation of such signatures. Some employed algorithms are fairly complex and hence sensitive to over-fitting whereas others are more simple and straight forward. Here we present a new type of simple algorithm based on ROC analysis and the use of metagenes that we believe will be a good complement to existing algorithms. Results The basis for the proposed approach is the use of metagenes, instead of collections of individual genes, and a feature selection using AUC values obtained by ROC analysis. Each gene in a data set is assigned an AUC value relative to the tumor class under investigation and the genes are ranked according to these values. Metagenes are then formed by calculating the mean expression level for an increasing number of ranked genes, and the metagene expression value that optimally discriminates tumor classes in the training set is used for classification of new samples. The performance of the metagene is then evaluated using LOOCV and balanced accuracies. Conclusions We show that the simple uni-variate gene expression average algorithm performs as well as several alternative algorithms such as discriminant analysis and the more complex approaches such as SVM and neural networks. The R package rocc is freely available at http://cran.r-project.org/web/packages/rocc/index.html.

Abstract Background The use of global gene expression profiling is a well established approach to understand biological processes. One of the major goals of these investigations is to identify sets of genes with similar expression patterns. Such gene signatures may be very informative and reveal new aspects of particular biological processes. A logical and systematic next step is to reduce the identified gene signatures to the regulatory components that induce the relevant gene expression changes. A central issue in this context is to identify transcription factors, or transcription factor binding sites (TFBS), likely to be of importance for the expression of the gene signatures. Results We develop a strategy that efficiently produces TFBS/promoter databases based on user-defined criteria. The resulting databases constitute all genes in the Santa Cruz database and the positions for all TFBS provided by the user as position weight matrices. These databases are then used for two purposes, to identify significant TFBS in the promoters in sets of genes and to identify clusters of co-occurring TFBS. We use two criteria for significance, significantly enriched TFBS in terms of total number of binding sites for the promoters, and significantly present TFBS in terms of the fraction of promoters with binding sites. Significant TFBS are identified by a re-sampling procedure in which the query gene set is compared with typically 105 gene lists of similar size randomly drawn from the TFBS/promoter database. We apply this strategy to a large number of published ChIP-Chip data sets and show that the proposed approach faithfully reproduces ChIP-Chip results. The strategy also identifies relevant TFBS when analyzing gene signatures obtained from the MSigDB database. In addition, we show that several TFBS are highly correlated and that co-occurring TFBS define functionally related sets of genes. Conclusions The presented approach of promoter analysis faithfully reproduces the results from several ChIP-Chip and MigDB derived gene sets and hence may prove to be an important method in the analysis of gene signatures obtained through ChIP-Chip or global gene expression experiments. We show that TFBS are organized in clusters of co-occurring TFBS that together define highly coherent sets of genes.

Abstract Background Mutations altering BLM function are associated with highly elevated cancer susceptibility (Bloom syndrome). Thus, genetic variants of BLM and proteins that form complexes with BLM, such as TOP3A and RMI1, might affect cancer risk as well. Methods In this study we have studied 26 tagged single nucleotide polymorphisms (tagSNPs) in RMI1, TOP3A, and BLM and their associations with cancer risk in acute myeloid leukemia/myelodysplatic syndromes (AML/MDS; N = 152), malignant melanoma (N = 170), and bladder cancer (N = 61). Two population-based control groups were used (N = 119 and N = 156). Results Based on consistency in effect estimates for the three cancer forms and similar allelic frequencies of the variant alleles in the control groups, two SNPs in TOP3A (rs1563634 and rs12945597) and two SNPs in BLM (rs401549 and rs2532105) were selected for analysis in breast cancer cases (N = 200) and a control group recruited from spouses of cancer patients (N = 131). The rs12945597 in TOP3A and rs2532105 in BLM showed increased risk for breast cancer. We then combined all cases (N = 584) and controls (N = 406) respectively and found significantly increased risk for variant carriers of rs1563634 A/G (AG carriers OR = 1.7 [95%CI 1.1–2.6], AA carriers OR = 1.8 [1.2–2.8]), rs12945597 G/A (GA carriers OR = 1.5 [1.1–1.9], AA carriers OR = 1.6 [1.0–2.5]), and rs2532105 C/T (CT+TT carriers OR = 1.8 [1.4–2.5]). Gene-gene interaction analysis suggested an additive effect of carrying more than one risk allele. For the variants of TOP3A, the risk increment was more pronounced for older carriers. Conclusion These results further support a role of low-penetrance genes involved in BLM-associated homologous recombination for cancer risk.

Abstract Background Illumina Infinium whole genome genotyping (WGG) arrays are increasingly being applied in cancer genomics to study gene copy number alterations and allele-specific aberrations such as loss-of-heterozygosity (LOH). Methods developed for normalization of WGG arrays have mostly focused on diploid, normal samples. However, for cancer samples genomic aberrations may confound normalization and data interpretation. Therefore, we examined the effects of the conventionally used normalization method for Illumina Infinium arrays when applied to cancer samples. Results We demonstrate an asymmetry in the detection of the two alleles for each SNP, which deleteriously influences both allelic proportions and copy number estimates. The asymmetry is caused by a remaining bias between the two dyes used in the Infinium II assay after using the normalization method in Illumina's proprietary software (BeadStudio). We propose a quantile normalization strategy for correction of this dye bias. We tested the normalization strategy using 535 individual hybridizations from 10 data sets from the analysis of cancer genomes and normal blood samples generated on Illumina Infinium II 300 k version 1 and 2, 370 k and 550 k BeadChips. We show that the proposed normalization strategy successfully removes asymmetry in estimates of both allelic proportions and copy numbers. Additionally, the normalization strategy reduces the technical variation for copy number estimates while retaining the response to copy number alterations. Conclusion The proposed normalization strategy represents a valuable tool that improves the quality of data obtained from Illumina Infinium arrays, in particular when used for LOH and copy number variation studies.

Abstract Background Urothelial carcinomas originate from the epithelial cells of the inner lining of the bladder and may appear as single or as multiple synchronous tumors. Patients with urothelial carcinomas frequently show recurrences after treatment making follow-up necessary. The leading hypothesis explaining the origin of meta- and synchronous tumors assumes a monoclonal origin. However, the genetic relationship among consecutive tumors has been shown to be complex in as much as the genetic evolution does not adhere to the chronological appearance of the metachronous tumors. Consequently, genetically less evolved tumors may appear chronologically later than genetically related but more evolved tumors. Methods Forty-nine meta- or synchronous urothelial tumors from 22 patients were analyzed using expression profiling, conventional CGH, LOH, and mutation analyses. Results We show by CGH that partial chromosomal losses in the initial tumors may not be present in the recurring tumors, by LOH that different haplotypes may be lost and that detected regions of LOH may be smaller in recurring tumors, and that mutations present in the initial tumor may not be present in the recurring ones. In contrast we show that despite apparent genomic differences, the recurrent and multiple bladder tumors from the same patients display remarkably similar expression profiles. Conclusion Our findings show that even though the vast majority of the analyzed meta- and synchronous tumors from the same patients are not likely to have originated directly from the preceding tumor they still show remarkably similar expressions profiles. The presented data suggests that an expression profile is established early in tumor development and that this profile is stable and maintained in recurring tumors.

Abstract Background Urothelial carcinoma (UC) is characterized by nonrandom chromosomal aberrations, varying from one or a few changes in early-stage and low-grade tumors, to highly rearranged karyotypes in muscle-invasive lesions. Recent array-CGH analyses have shed further light on the genomic changes underlying the neoplastic development of UC, and have facilitated the molecular delineation amplified and deleted regions to the level of specific candidate genes. In the present investigation we combine detailed genomic information with expression information to identify putative target genes for genomic amplifications. Methods We analyzed 38 urothelial carcinomas by whole-genome tiling resolution array-CGH and high density expression profiling to identify putative target genes in common genomic amplifications. When necessary expression profiling was complemented with Q-PCR of individual genes. Results Three genomic segments were frequently and exclusively amplified in high grade tumors; 1q23, 6p22 and 8q22, respectively. Detailed mapping of the 1q23 segment showed a heterogeneous amplification pattern and no obvious commonly amplified region. The 6p22 amplicon was defined by a 1.8 Mb core region present in all amplifications, flanked both distally and proximally by segments amplified to a lesser extent. By combining genomic profiles with expression profiles we could show that amplification of E2F3, CDKAL1, SOX4, and MBOAT1 as well as NUP153, AOF1, FAM8A1 and DEK in 6p22 was associated with increased gene expression. Amplification of the 8q22 segment was primarily associated with YWHAZ (14-3-3-zeta) and POLR2K over expression. The possible importance of the YWHA genes in the development of urothelial carcinomas was supported by another recurrent amplicon paralogous to 8q22, in 2p25, where increased copy numbers lead to enhanced expression of YWHAQ (14-3-3-theta). Homozygous deletions were identified at 10 different genomic locations, most frequently affecting CDKN2A/CDKN2B in 9p21 (32%). Notably, the latter occurred mutually exclusive with 6p22 amplifications. Conclusion The presented data indicates 6p22 as a composite amplicon with more than one possible target gene. The data also suggests that amplification of 6p22 and homozygous deletions of 9p21 may have complementary roles. Furthermore, the analysis of paralogous regions that showed genomic amplification indicated altered expression of YWHA (14-3-3) genes as important events in the development of UC.

Abstract Background The use of global gene expression profiling to identify sets of genes with similar expression patterns is rapidly becoming a widespread approach for understanding biological processes. A logical and systematic approach to study co-expressed genes is to analyze their promoter sequences to identify transcription factors that may be involved in establishing specific profiles and that may be experimentally investigated. Results We introduce promoter clustering i.e. grouping of promoters with respect to their high scoring motif content, and show that this approach greatly enhances the identification of common and significant transcription factor binding sites (TFBS) in co-expressed genes. We apply this method to two different dataset, one consisting of micro array data from 108 leukemias (AMLs) and a second from a time series experiment, and show that biologically relevant promoter patterns may be obtained using phylogenetic foot-printing methodology. In addition, we also found that 15% of the analyzed promoter regions contained transcription factors start sites for additional genes transcribed in the opposite direction. Conclusion Promoter clustering based on global promoter features greatly improve the identification of shared TFBS in co-expressed genes. We believe that the outlined approach may be a useful first step to identify transcription factors that contribute to specific features of gene expression profiles.

Abstract Background An alternative to standard approaches to uncover biologically meaningful structures in micro array data is to treat the data as a blind source separation (BSS) problem. BSS attempts to separate a mixture of signals into their different sources and refers to the problem of recovering signals from several observed linear mixtures. In the context of micro array data, "sources" may correspond to specific cellular responses or to co-regulated genes. Results We applied independent component analysis (ICA) to three different microarray data sets; two tumor data sets and one time series experiment. To obtain reliable components we used iterated ICA to estimate component centrotypes. We found that many of the low ranking components indeed may show a strong biological coherence and hence be of biological significance. Generally ICA achieved a higher resolution when compared with results based on correlated expression and a larger number of gene clusters with significantly enriched for gene ontology (GO) categories. In addition, components characteristic for molecular subtypes and for tumors with specific chromosomal translocations were identified. ICA also identified more than one gene clusters significant for the same GO categories and hence disclosed a higher level of biological heterogeneity, even within coherent groups of genes. Conclusion Although the ICA approach primarily detects hidden variables, these surfaced as highly correlated genes in time series data and in one instance in the tumor data. This further strengthens the biological relevance of latent variables detected by ICA.

Therapeutic antibodies that block the programmed death-1 (PD-1)-programmed death-ligand 1 (PD-L1) pathway can induce robust and durable responses in patients with various cancers, including metastatic urothelial cancer. However, these responses only occur in a subset of patients. Elucidating the determinants of response and resistance is key to improving outcomes and developing new treatment strategies. Here we examined tumours from a large cohort of patients with metastatic urothelial cancer who were treated with an anti-PD-L1 agent (atezolizumab) and identified major determinants of clinical outcome. Response to treatment was associated with CD8+ T-effector cell phenotype and, to an even greater extent, high neoantigen or tumour mutation burden. Lack of response was associated with a signature of transforming growth factor β (TGFβ) signalling in fibroblasts. This occurred particularly in patients with tumours, which showed exclusion of CD8+ T cells from the tumour parenchyma that were instead found in the fibroblast- and collagen-rich peritumoural stroma; a common phenotype among patients with metastatic urothelial cancer. Using a mouse model that recapitulates this immune-excluded phenotype, we found that therapeutic co-administration of TGFβ-blocking and anti-PD-L1 antibodies reduced TGFβ signalling in stromal cells, facilitated T-cell penetration into the centre of tumours, and provoked vigorous anti-tumour immunity and tumour regression. Integration of these three independent biological features provides the best basis for understanding patient outcome in this setting and suggests that TGFβ shapes the tumour microenvironment to restrain anti-tumour immunity by restricting T-cell infiltration.

Post-traumatic stress disorder (PTSD) is a heterogeneous condition evidenced by the absence of objective physiological measurements applicable to all who meet the criteria for the disorder as well as divergent responses to treatments. This study capitalized on biological diversity observed within the PTSD group observed following epigenome-wide analysis of a well-characterized Discovery cohort (N = 166) consisting of 83 male combat exposed veterans with PTSD, and 83 combat veterans without PTSD in order to identify patterns that might distinguish subtypes. Computational analysis of DNA methylation (DNAm) profiles identified two PTSD biotypes within the PTSD+ group, G1 and G2, associated with 34 clinical features that are associated with PTSD and PTSD comorbidities. The G2 biotype was associated with an increased PTSD risk and had higher polygenic risk scores and a greater methylation compared to the G1 biotype and healthy controls. The findings were validated at a 3-year follow-up (N = 59) of the same individuals as well as in two independent, veteran cohorts (N = 54 and N = 38), and an active duty cohort (N = 133). In some cases, for example Dopamine-PKA-CREB and GABA-PKC-CREB signaling pathways, the biotypes were oppositely dysregulated, suggesting that the biotypes were not simply a function of a dimensional relationship with symptom severity, but may represent distinct biological risk profiles underpinning PTSD. The identification of two novel distinct epigenetic biotypes for PTSD may have future utility in understanding biological and clinical heterogeneity in PTSD and potential applications in risk assessment for active duty military personnel under non-clinician-administered settings, and improvement of PTSD diagnostic markers.

Urothelial cancer of the urinary bladder frequently metastasizes to lymph‐nodes, lungs, liver and bone. A taxonomy for molecular classification exists, but it is unknown if molecular subtypes show tropism for different organs. Here, we study 146 patients with de novo metastatic disease or recurrence after curative treatment. We classify primary tumors using two transcriptomic methods and immunostaining and identify enrichment and depletion of metastatic sites in molecular subtypes using permutation tests. We observed significant depletion of bone metastases in the Basal/squamous molecular subtype, whereas the Urothelial‐like subtype entailed an enrichment for metastases to bone. The Genomically unstable subtype was depleted of lung metastases, but enriched for atypical sites, including six out of seven patients with brain metastases. Stroma‐rich primary tumor samples were associated with local recurrence, but not with distant sites. Additionally, the proportion with brain or testis metastases differed between systemic chemotherapy regimens (GC vs MVAC) suggesting a sanctuary effect. In conclusion, molecular subtypes of urothelial bladder cancer are significantly associated with specific metastatic sites, suggesting that subtype‐specific molecular determinants could exist at various steps in the metastatic cascade. [ABSTRACT FROM AUTHOR]

Breast tumors of the basal-like, hormone receptor-negative subtype remain an unmet clinical challenge, as there is high rate of recurrence and poor survival in patients following treatment. Coevolution of the malignant mammary epithelium and its underlying stroma instigates cancer-associated fibroblasts (CAFs) to support most, if not all, hallmarks of cancer progression. Here we delineate a previously unappreciated role for CAFs as determinants of the molecular subtype of breast cancer. We identified paracrine crosstalk between cancer cells expressing platelet-derived growth factor (PDGF)-CC and CAFs expressing the cognate receptors in human basal-like mammary carcinomas. Genetic or pharmacological intervention of PDGF-CC activity in mouse models of cancer resulted in conversion of basal-like breast cancers into a hormone receptor-positive state that enhanced sensitivity to endocrine therapy in previously resistant tumors. We conclude that specification of breast cancer to the basal-like subtype is under microenvironmental control and is therapeutically actionable., Author(s): Pernilla Roswall [1, 2]; Matteo Bocci [1]; Michael Bartoschek [1]; Hong Li [2]; Glen Kristiansen [3]; Sara Jansson [4]; Sophie Lehn [1]; Jonas Sjölund [1]; Steven Reid [1]; Christer [...]