Detection of treatment-induced changes in signaling pathways in gastrointestinal stromal tumors using transcriptomic data

Targeting BCL6 in Gastrointestinal Stromal Tumor Promotes p53-Mediated Apoptosis to Enhance the Antitumor Activity of Imatinib

Imatinib mesylate (IM) has revolutionized the treatment of gastrointestinal stromal tumor (GIST). However, most patients inevitably acquire IM resistance. Second- and third-line treatments exhibit modest clinical benefits with a median time to disease progression of 4 to 6 months, highlighting the urgency for novel therapeutic approaches. Here, we report that the expression of BCL6, a known oncogenic driver and transcriptional repressor, was significantly induced in GIST cells following IM treatment. Elevated BCL6 levels suppressed apoptosis and contributed to IM resistance. Mechanistically, BCL6 recruited SIRT1 to the TP53 promoter to modulate histone acetylation and transcriptionally repress TP53 expression. The reduction in p53 subsequently attenuated cell apoptosis and promoted tolerance of GIST cells to IM. Concordantly, treatment of GIST cells showing high BCL6 expression with a BCL6 inhibitor, BI-3802, conferred IM sensitivity. Furthermore, BI-3802 showed striking synergy with IM in IM-responsive and IM-resistant GIST cells in vitro and in vivo. Thus, these findings reveal a role for BCL6 in IM resistance and suggest that a combination of BCL6 inhibitors and IM could be a potentially effective treatment for GIST.

SIGNIFICANCE

BCL6 drives resistance to imatinib by inhibiting p53-mediated apoptosis and can be targeted in combination with imatinib to synergistically suppress tumor growth, providing a therapeutic strategy for treating gastrointestinal stromal tumor.

Metabolomic and transcriptomic response to imatinib treatment of gastrointestinal stromal tumour in xenograft-bearing mice

BACKGROUND

Although imatinib is a well-established first-line drug for treating a vast majority of gastrointestinal stromal tumours (GIST), GISTs acquire secondary resistance during therapy. Multi-omics approaches provide an integrated perspective to empower the development of personalised therapies through a better understanding of functional biology underlying the disease and molecular-driven selection of the best-targeted individualised therapy. In this study, we applied integrative metabolomic and transcriptomic analyses to elucidate tumour biochemical processes affected by imatinib treatment.

MATERIALS AND METHODS

A GIST xenograft mouse model was used in the study, including 10 mice treated with imatinib and 10 non-treated controls. Metabolites in tumour extracts were analysed using gas chromatography coupled with mass spectrometry (GC-MS). RNA sequencing was also performed on the samples subset (n=6).

RESULTS

Metabolomic analysis revealed 21 differentiating metabolites, whereas next-generation RNA sequencing data analysis resulted in 531 differentially expressed genes. Imatinib significantly changed the profile of metabolites associated mainly with purine and pyrimidine metabolism, butanoate metabolism, as well as alanine, aspartate, and glutamate metabolism. The related changes in transcriptomic profiles included genes involved in kinase activity and immune responses, as well as supported its impact on the purine biosynthesis pathway.

CONCLUSIONS

Our multi-omics study confirmed previously known pathways involved in imatinib anticancer activity as well as correlated imatinib-relevant downregulation of expression of purine biosynthesis pathway genes with the reduction of respectful metabolites. Furthermore, considering the importance of the purine biosynthesis pathway for cancer proliferation, we identified a potentially novel mechanism for the anti-tumour activity of imatinib. Based on the results, we hypothesise metabolic modulations aiming at the reduction in purine and pyrimidine pool may ensure higher imatinib efficacy or re-sensitise imatinib-resistant tumours.

Transfer learning between preclinical models and human tumors identifies a conserved NK cell activation signature in anti-CTLA-4 responsive tumors

BACKGROUND

Tumor response to therapy is affected by both the cell types and the cell states present in the tumor microenvironment. This is true for many cancer treatments, including immune checkpoint inhibitors (ICIs). While it is well-established that ICIs promote T cell activation, their broader impact on other intratumoral immune cells is unclear; this information is needed to identify new mechanisms of action and improve ICI efficacy. Many preclinical studies have begun using single-cell analysis to delineate therapeutic responses in individual immune cell types within tumors. One major limitation to this approach is that therapeutic mechanisms identified in preclinical models have failed to fully translate to human disease, restraining efforts to improve ICI efficacy in translational research.

METHOD

We previously developed a computational transfer learning approach called projectR to identify shared biology between independent high-throughput single-cell RNA-sequencing (scRNA-seq) datasets. In the present study, we test this algorithm's ability to identify conserved and clinically relevant transcriptional changes in complex tumor scRNA-seq data and expand its application to the comparison of scRNA-seq datasets with additional data types such as bulk RNA-seq and mass cytometry.

RESULTS

We found a conserved signature of NK cell activation in anti-CTLA-4 responsive mouse and human tumors. In human metastatic melanoma, we found that the NK cell activation signature associates with longer overall survival and is predictive of anti-CTLA-4 (ipilimumab) response. Additional molecular approaches to confirm the computational findings demonstrated that human NK cells express CTLA-4 and bind anti-CTLA-4 antibodies independent of the antibody binding receptor (FcR) and that similar to T cells, CTLA-4 expression by NK cells is modified by cytokine-mediated and target cell-mediated NK cell activation.

CONCLUSIONS

These data demonstrate a novel application of our transfer learning approach, which was able to identify cell state transitions conserved in preclinical models and human tumors. This approach can be adapted to explore many questions in cancer therapeutics, enhance translational research, and enable better understanding and treatment of disease.

Decomposing Cell Identity for Transfer Learning across Cellular Measurements, Platforms, Tissues, and Species

Analysis of gene expression in single cells allows for decomposition of cellular states as low-dimensional latent spaces. However, the interpretation and validation of these spaces remains a challenge. Here, we present scCoGAPS, which defines latent spaces from a source single-cell RNA-sequencing (scRNA-seq) dataset, and projectR, which evaluates these latent spaces in independent target datasets via transfer learning. Application of developing mouse retina to scRNA-Seq reveals intrinsic relationships across biological contexts and assays while avoiding batch effects and other technical features. We compare the dimensions learned in this source dataset to adult mouse retina, a time-course of human retinal development, select scRNA-seq datasets from developing brain, chromatin accessibility data, and a murine-cell type atlas to identify shared biological features. These tools lay the groundwork for exploratory analysis of scRNA-seq data via latent space representations, enabling a shift in how we compare and identify cells beyond reliance on marker genes or ensemble molecular identity.

DeepMF: deciphering the latent patterns in omics profiles with a deep learning method

BACKGROUND

With recent advances in high-throughput technologies, matrix factorization techniques are increasingly being utilized for mapping quantitative omics profiling matrix data into low-dimensional embedding space, in the hope of uncovering insights in the underlying biological processes. Nevertheless, current matrix factorization tools fall short in handling noisy data and missing entries, both deficiencies that are often found in real-life data.

RESULTS

Here, we propose DeepMF, a deep neural network-based factorization model. DeepMF disentangles the association between molecular feature-associated and sample-associated latent matrices, and is tolerant to noisy and missing values. It exhibited feasible cancer subtype discovery efficacy on mRNA, miRNA, and protein profiles of medulloblastoma cancer, leukemia cancer, breast cancer, and small-blue-round-cell cancer, achieving the highest clustering accuracy of 76%, 100%, 92%, and 100% respectively. When analyzing data sets with 70% missing entries, DeepMF gave the best recovery capacity with silhouette values of 0.47, 0.6, 0.28, and 0.44, outperforming other state-of-the-art MF tools on the cancer data sets Medulloblastoma, Leukemia, TCGA BRCA, and SRBCT. Its embedding strength as measured by clustering accuracy is 88%, 100%, 84%, and 96% on these data sets, which improves on the current best methods 76%, 100%, 78%, and 87%.

CONCLUSION

DeepMF demonstrated robust denoising, imputation, and embedding ability. It offers insights to uncover the underlying biological processes such as cancer subtype discovery. Our implementation of DeepMF can be found at https://github.com/paprikachan/DeepMF.

Enter the Matrix: Factorization Uncovers Knowledge from Omics

Omics data contain signals from the molecular, physical, and kinetic inter- and intracellular interactions that control biological systems. Matrix factorization (MF) techniques can reveal low-dimensional structure from high-dimensional data that reflect these interactions. These techniques can uncover new biological knowledge from diverse high-throughput omics data in applications ranging from pathway discovery to timecourse analysis. We review exemplary applications of MF for systems-level analyses. We discuss appropriate applications of these methods, their limitations, and focus on the analysis of results to facilitate optimal biological interpretation. The inference of biologically relevant features with MF enables discovery from high-throughput data beyond the limits of current biological knowledge - answering questions from high-dimensional data that we have not yet thought to ask.

Integrated time course omics analysis distinguishes immediate therapeutic response from acquired resistance

BACKGROUND

Targeted therapies specifically act by blocking the activity of proteins that are encoded by genes critical for tumorigenesis. However, most cancers acquire resistance and long-term disease remission is rarely observed. Understanding the time course of molecular changes responsible for the development of acquired resistance could enable optimization of patients' treatment options. Clinically, acquired therapeutic resistance can only be studied at a single time point in resistant tumors.

METHODS

To determine the dynamics of these molecular changes, we obtained high throughput omics data (RNA-sequencing and DNA methylation) weekly during the development of cetuximab resistance in a head and neck cancer in vitro model. The CoGAPS unsupervised algorithm was used to determine the dynamics of the molecular changes associated with resistance during the time course of resistance development.

RESULTS

CoGAPS was used to quantify the evolving transcriptional and epigenetic changes. Applying a PatternMarker statistic to the results from CoGAPS enabled novel heatmap-based visualization of the dynamics in these time course omics data. We demonstrate that transcriptional changes result from immediate therapeutic response or resistance, whereas epigenetic alterations only occur with resistance. Integrated analysis demonstrates delayed onset of changes in DNA methylation relative to transcription, suggesting that resistance is stabilized epigenetically.

CONCLUSIONS

Genes with epigenetic alterations associated with resistance that have concordant expression changes are hypothesized to stabilize the resistant phenotype. These genes include FGFR1, which was associated with EGFR inhibitors resistance previously. Thus, integrated omics analysis distinguishes the timing of molecular drivers of resistance. This understanding of the time course progression of molecular changes in acquired resistance is important for the development of alternative treatment strategies that would introduce appropriate selection of new drugs to treat cancer before the resistant phenotype develops.

PatternMarkers & GWCoGAPS for novel data-driven biomarkers via whole transcriptome NMF

Summary

Non-negative Matrix Factorization (NMF) algorithms associate gene expression with biological processes (e.g. time-course dynamics or disease subtypes). Compared with univariate associations, the relative weights of NMF solutions can obscure biomarkers. Therefore, we developed a novel patternMarkers statistic to extract genes for biological validation and enhanced visualization of NMF results. Finding novel and unbiased gene markers with patternMarkers requires whole-genome data. Therefore, we also developed Genome-Wide CoGAPS Analysis in Parallel Sets (GWCoGAPS), the first robust whole genome Bayesian NMF using the sparse, MCMC algorithm, CoGAPS. Additionally, a manual version of the GWCoGAPS algorithm contains analytic and visualization tools including patternMatcher, a Shiny web application. The decomposition in the manual pipeline can be replaced with any NMF algorithm, for further generalization of the software. Using these tools, we find granular brain-region and cell-type specific signatures with corresponding biomarkers in GTEx data, illustrating GWCoGAPS and patternMarkers ascertainment of data-driven biomarkers from whole-genome data.

Availability and Implementation

PatternMarkers & GWCoGAPS are in the CoGAPS Bioconductor package (3.5) under the GPL license.

Contact

gsteinobrien@jhmi.edu or ccolantu@jhmi.edu or ejfertig@jhmi.edu.

Supplementary information

Supplementary data are available at Bioinformatics online.

CoGAPS matrix factorization algorithm identifies transcriptional changes in AP-2alpha target genes in feedback from therapeutic inhibition of the EGFR network

Patients with oncogene driven tumors are treated with targeted therapeutics including EGFR inhibitors. Genomic data from The Cancer Genome Atlas (TCGA) demonstrates molecular alterations to EGFR, MAPK, and PI3K pathways in previously untreated tumors. Therefore, this study uses bioinformatics algorithms to delineate interactions resulting from EGFR inhibitor use in cancer cells with these genetic alterations. We modify the HaCaT keratinocyte cell line model to simulate cancer cells with constitutive activation of EGFR, HRAS, and PI3K in a controlled genetic background. We then measure gene expression after treating modified HaCaT cells with gefitinib, afatinib, and cetuximab. The CoGAPS algorithm distinguishes a gene expression signature associated with the anticipated silencing of the EGFR network. It also infers a feedback signature with EGFR gene expression itself increasing in cells that are responsive to EGFR inhibitors. This feedback signature has increased expression of several growth factor receptors regulated by the AP-2 family of transcription factors. The gene expression signatures for AP-2alpha are further correlated with sensitivity to cetuximab treatment in HNSCC cell lines and changes in EGFR expression in HNSCC tumors with low CDKN2A gene expression. In addition, the AP-2alpha gene expression signatures are also associated with inhibition of MEK, PI3K, and mTOR pathways in the Library of Integrated Network-Based Cellular Signatures (LINCS) data. These results suggest that AP-2 transcription factors are activated as feedback from EGFR network inhibition and may mediate EGFR inhibitor resistance.

Influence of selected anti-cancer drugs on the induction of DNA double-strand breaks and changes in gene expression in human hepatoma HepG2 cells

In chemotherapy, various anti-cancer drugs with different mechanisms of action are used and may represent different risk of undesirable delayed side effects in treated patients as well as in occupationally exposed populations. The aim of the present study was to evaluate genotoxic potential of four widely used anti-cancer drugs with different mechanisms of action: 5-fluorouracil (5-FU), cisplatin (CDDP) and etoposide (ET) that cause cell death by targeting DNA function and imatinib mesylate (IM) that inhibits targeted protein kinases in cancer cells in an experimental model with human hepatoma HepG2 cells. After 24 h of exposure all four anti-cancer drugs at non-cytotoxic concentrations induced significant increase in formation of DNA double strand breaks (DSBs), with IM being the least effective. The analysis of the changes in the expression of genes involved in the response to DNA damage (CDKN1A, GADD45A, MDM2), apoptosis (BAX, BCL2) and oncogenesis (MYC, JUN) showed that 5-FU, CDDP and ET upregulated the genes involved in DNA damage response, while the anti-apoptotic gene BCL2 and oncogene MYC were downregulated. On the contrary, IM did not change the mRNA level of the studied genes, showing different mechanism of action that probably does not involve direct interaction with DNA processing. Genotoxic effects of the tested anti-cancer drugs were observed at their therapeutic concentrations that may consequently lead to increased risk for development of delayed adverse effects in patients. In addition, considering the genotoxic mechanism of action of 5-FU, CDDP and ET an increased risk can also not be excluded in occupationally exposed populations. The results also indicate that exposure to 5-FU, CDDP and ET represent a higher risk for delayed effects such as cancer, reproductive effects and heritable disease than exposure to IM.

Toward Signaling-Driven Biomarkers Immune to Normal Tissue Contamination

The goal of this study was to discover a minimally invasive pathway-specific biomarker that is immune to normal cell mRNA contamination for diagnosing head and neck squamous cell carcinoma (HNSCC). Using Elsevier's MedScan natural language processing component of the Pathway Studio software and the TRANSFAC database, we produced a curated set of genes regulated by the signaling networks driving the development of HNSCC. The network and its gene targets provided prior probabilities for gene expression, which guided our CoGAPS matrix factorization algorithm to isolate patterns related to HNSCC signaling activity from a microarray-based study. Using patterns that distinguished normal from tumor samples, we identified a reduced set of genes to analyze with Top Scoring Pair in order to produce a potential biomarker for HNSCC. Our proposed biomarker comprises targets of the transcription factor (TF) HIF1A and the FOXO family of TFs coupled with genes that show remarkable stability across all normal tissues. Based on validation with novel data from The Cancer Genome Atlas (TCGA), measured by RNAseq, and bootstrap sampling, the biomarker for normal vs. tumor has an accuracy of 0.77, a Matthews correlation coefficient of 0.54, and an area under the curve (AUC) of 0.82.

Beyond standard therapy: drugs under investigation for the treatment of gastrointestinal stromal tumor

INTRODUCTION

Gastrointestinal stromal tumor (GIST) is the most common nonepithelial malignancy of the GI tract. With the discovery of KIT and later platelet-derived growth factor α (PDGFRA) gain-of-function mutations as factors in the pathogenesis of the disease, GIST was the quintessential model for targeted therapy. Despite the successful clinical use of imatinib mesylate, a selective receptor tyrosine kinase (RTK) inhibitor that targets KIT, PDGFRA and BCR-ABL, we still do not have treatment for the long-term control of advanced GIST.

AREAS COVERED

This review summarizes the drugs that are under investigation or have been assessed in trials for GIST treatment. The article focuses on their mechanisms of actions, the preclinical evidence of efficacy, and the clinical trials concerning safety and efficacy in humans.

EXPERT OPINION

It is known that KIT and PDGFRA mutations in GIST patients influence the response to treatment. This observation should be taken into consideration when investigating new drugs. RECIST was developed to help uniformly report efficacy trials in oncology. Despite the usefulness of this system, many questions are being addressed about its validity in evaluating the true efficacy of drugs knowing that new targeted therapies do not affect the tumor size as much as they halt progression and prolong survival.

Correcting transcription factor gene sets for copy number and promoter methylation variations

Gene set analysis provides a method to generate statistical inferences across sets of linked genes, primarily using high-throughput expression data. Common gene sets include biological pathways, operons, and targets of transcriptional regulators. In higher eukaryotes, especially when dealing with diseases with strong genetic and epigenetic components such as cancer, copy number loss and gene silencing through promoter methylation can eliminate the possibility that a gene is transcribed. This, in turn, can adversely affect the estimation of transcription factor or pathway activity from a set of target genes, as some of the targets may not be responsive to transcriptional regulation. Here we introduce a simple filtering approach that removes genes from consideration if they show copy number loss or promoter methylation, and demonstrate the improvement in inference of transcription factor activity in a simulated dataset based on the background expression observed in normal head and neck tissue.

An argument for mechanism-based statistical inference in cancer

Cancer is perhaps the prototypical systems disease, and as such has been the focus of extensive study in quantitative systems biology. However, translating these programs into personalized clinical care remains elusive and incomplete. In this perspective, we argue that realizing this agenda—in particular, predicting disease phenotypes, progression and treatment response for individuals—requires going well beyond standard computational and bioinformatics tools and algorithms. It entails designing global mathematical models over network-scale configurations of genomic states and molecular concentrations, and learning the model parameters from limited available samples of high-dimensional and integrative omics data. As such, any plausible design should accommodate: biological mechanism, necessary for both feasible learning and interpretable decision making; stochasticity, to deal with uncertainty and observed variation at many scales; and a capacity for statistical inference at the patient level. This program, which requires a close, sustained collaboration between mathematicians and biologists, is illustrated in several contexts, including learning biomarkers, metabolism, cell signaling, network inference and tumorigenesis.

A Kinetic-Model-Based Approach to Identify Malfunctioning Components in Signal Transduction Pathways from Artificial Clinical Data

Detection of malfunctioning reactions or molecules from clinical data is essential for disease treatments. In order to find an alternative to the existing oversimplistic mathematical models, a kinetic model is developed in this work to infer the malfunctioning reactions/molecules by quantifying the similarity between the clinical profile and the output profiles predicted from the model in which certain reactions/molecules malfunction. The new approach was tested in IL-6 and TNF-α/NF-κB signaling pathway, for four abnormal conditions including up/downregulation of single reaction rate constants and up/downregulation of single molecules. Since limited quantitative clinical data were available, the IL-6 ODE model was used to generate artificial clinical data for the abnormal steady-state value shown in two key molecules: nuclear STAT3 and SOCS3. Similarly, the TNF-α/NF-κB model was used to obtain the data in which abnormal oscillation dynamic was shown in the profile of NF-κB. The results show that the approach developed in this study was able to successfully identify the malfunctioning reactions and molecules from the clinical data. It was also found that this new approach was noise-robust and that it managed to reveal unique solution for the faulty components in a network.

Learning dysregulated pathways in cancers from differential variability analysis

Analysis of gene sets can implicate activity in signaling pathways that is responsible for cancer initiation and progression, but is not discernible from the analysis of individual genes. Multiple methods and software packages have been developed to infer pathway activity from expression measurements for set of genes targeted by that pathway. Broadly, three major methodologies have been proposed: over-representation, enrichment, and differential variability. Both over-representation and enrichment analyses are effective techniques to infer differentially regulated pathways from gene sets with relatively consistent differentially expressed (DE) genes. Specifically, these algorithms aggregate statistics from each gene in the pathway. However, they overlook multivariate patterns related to gene interactions and variations in expression. Therefore, the analysis of differential variability of multigene expression patterns can be essential to pathway inference in cancers. The corresponding methodologies and software packages for such multivariate variability analysis of pathways are reviewed here. We also introduce a new, computationally efficient algorithm, expression variation analysis (EVA), which has been implemented along with a previously proposed algorithm, Differential Rank Conservation (DIRAC), in an open source R package, gene set regulation (GSReg). EVA inferred similar pathways as DIRAC at reduced computational costs. Moreover, EVA also inferred different dysregulated pathways than those identified by enrichment analysis.

Drug repurposing identifies a synergistic combination therapy with imatinib mesylate for gastrointestinal stromal tumor

Gastrointestinal stromal tumor (GIST) is a rare and therefore often neglected disease. Introduction of the kinase inhibitor imatinib mesylate radically improved the clinical response of patients with GIST; however, its effects are often short-lived, with GISTs demonstrating a median time-to-progression of approximately two years. Although many investigational drugs, approved first for other cancers, have been subsequently evaluated for the management of GIST, few have greatly affected the overall survival of patients with advanced disease. We employed a novel, focused, drug-repurposing effort for GIST, including imatinib mesylate-resistant GIST, evaluating a large library of FDA-approved drugs regardless of current indication. As a result of the drug-repurposing screen, we identified eight FDA-approved drugs, including fludarabine phosphate (F-AMP), that showed synergy with and/or overcame resistance to imatinib mesylate. F-AMP induces DNA damage, Annexin V, and caspase-3/7 activities as the cytotoxic effects on GIST cells, including imatinib mesylate-resistant GIST cells. F-AMP and imatinib mesylate combination treatment showed greater inhibition of GIST cell proliferation when compared with imatinib mesylate and F-AMP alone. Successful in vivo experiments confirmed the combination of imatinib mesylate with F-AMP enhanced the antitumor effects compared with imatinib mesylate alone. Our results identified F-AMP as a promising, repurposed drug therapy for the treatment of GISTs, with potential to be administered in combination with imatinib mesylate or for treatment of imatinib mesylate-refractory tumors.

Outlier Analysis and Top Scoring Pair for Integrated Data Analysis and Biomarker Discovery

Pathway deregulation has been identified as a key driver of carcinogenesis, with proteins in signaling pathways serving as primary targets for drug development. Deregulation can be driven by a number of molecular events, including gene mutation, epigenetic changes in gene promoters, overexpression, and gene amplifications or deletions. We demonstrate a novel approach that identifies pathways of interest by integrating outlier analysis within and across molecular data types with gene set analysis. We use the results to seed the top-scoring pair algorithm to identify robust biomarkers associated with pathway deregulation. We demonstrate this methodology on pediatric acute myeloid leukemia (AML) data. We develop a biomarker in primary AML tumors, demonstrate robustness with an independent primary tumor data set, and show that the identified biomarkers also function well in relapsed pediatric AML tumors.

Tumor-derived exosomes: A message delivery system for tumor progression

Intercellular communication is a key process in the development and progression of cancer. The dynamic and reciprocal interplays between the tumor and its microenvironment orchestrate events critical to the establishment of primary and metastatic niches and maintenance of a permissive environment at the tumor−stroma interface. Atay and colleagues found that gastrointestinal stromal tumor cells secrete vesicles known as exosomes. These exosomes contain oncogenic KIT and their transfer and uptake by surrounding smooth muscle cells lead to enhanced AKT and MAPK signaling and phenotypic modulation of several cellular processes, including morphological changes, expression of tumor-associated markers, secretion of matrix metalloproteinases, and enhanced tumor cell invasion. This provocative study emphasizes that exosome-mediated signaling within the tumor microenvironment acts as a positive feedback loop that contributes to invasiveness and that interfering with this message delivery system may represent promising therapeutic approaches, not only for GIST, but for other types of cancer.

Pattern identification in time-course gene expression data with the CoGAPS matrix factorization

Patterns in time-course gene expression data can represent the biological processes that are active over the measured time period. However, the orthogonality constraint in standard pattern-finding algorithms, including notably principal components analysis (PCA), confounds expression changes resulting from simultaneous, non-orthogonal biological processes. Previously, we have shown that Markov chain Monte Carlo nonnegative matrix factorization algorithms are particularly adept at distinguishing such concurrent patterns. One such matrix factorization is implemented in the software package CoGAPS. We describe the application of this software and several technical considerations for identification of age-related patterns in a public, prefrontal cortex gene expression dataset.

Preferential activation of the hedgehog pathway by epigenetic modulations in HPV negative HNSCC identified with meta-pathway analysis

Head and neck squamous cell carcinoma (HNSCC) is largely divided into two groups based on their etiology, human papillomavirus (HPV)-positive and –negative. Global DNA methylation changes are known to drive oncogene and tumor suppressor expression in primary HNSCC of both types. However, significant heterogeneity in DNA methylation within the groups results in different transcriptional profiles and clinical outcomes. We applied a meta-pathway analysis to link gene expression changes to DNA methylation in distinguishing HNSCC subtypes. This approach isolated specific epigenetic changes controlling expression in HPV− HNSCC that distinguish it from HPV+ HNSCC. Analysis of genes identified Hedgehog pathway activation specific to HPV− HNSCC. We confirmed that GLI1, the primary Hedgehog target, showed higher expression in tumors compared to normal samples with HPV− tumors having the highest GLI1 expression, suggesting that increased expression of GLI1 is a potential driver in HPV− HNSCC. Our algorithm for integration of DNA methylation and gene expression can infer biologically significant molecular pathways that may be exploited as therapeutics targets. Our results suggest that therapeutics targeting the Hedgehog pathway may be of benefit in HPV− HNSCC. Similar integrative analysis of high-throughput coupled DNA methylation and expression datasets may yield novel insights into deregulated pathways in other cancers.

Capturing drug responses by quantitative promoter activity profiling

Quantitative analysis of cellular responses to drugs is of major interest in pharmaceutical research. Microarray technologies have been widely used for monitoring genome-wide expression changes. However, this approach has several limitations in terms of coverage of targeted RNAs, sensitivity, and quantitativeness, which are crucial for accurate monitoring of cellular responses. In this article, we report an application of genome-wide and quantitative profiling of cellular responses to drugs. We monitored promoter activities in MCF-7 cells by Cap Analysis of Gene Expression using a single-molecule sequencer. We identified a distinct set of promoters affected even by subtle inhibition of the Ras-ERK and phosphatidylinositol-3-kinase-Akt signal-transduction pathways. Furthermore, we succeeded in explaining the majority of promoter responses to inhibition of the upstream epidermal growth factor receptor kinase quantitatively based on the promoter profiles upon inhibition of the two individual downstream signaling pathways. Our results demonstrate unexplored utility of highly quantitative promoter activity profiling in drug research.

Identifying context-specific transcription factor targets from prior knowledge and gene expression data

Numerous methodologies, assays, and databases presently provide candidate targets of transcription factors (TFs). However, TFs rarely regulate their targets universally. The context of activation of a TF can change the transcriptional response of targets. Direct multiple regulation typical to mammalian genes complicates direct inference of TF targets from gene expression data. We present a novel statistic that infers context-specific TF regulation based upon the CoGAPS algorithm, which infers overlapping gene expression patterns resulting from coregulation. Numerical experiments with simulated data showed that this statistic correctly inferred targets that are common to multiple TFs, except in cases where the signal from a TF is negligible relative to noise level and signal from other TFs. The statistic is robust to moderate levels of error in the simulated gene sets, identifying fewer false positives than false negatives. Significantly, the regulatory statistic refines the number of TF targets relevant to cell signaling in gastrointestinal stromal tumors (GIST) to genes consistent with the phosphorylation patterns of TFs identified in previous studies. As formulated, the proposed regulatory statistic has wide applicability to inferring set membership in integrated datasets. This statistic could be naturally extended to account for prior probabilities of set membership or to add candidate gene targets.

The non-negative matrix factorization toolbox for biological data mining

Background

Non-negative matrix factorization (NMF) has been introduced as an important method for mining biological data. Though there currently exists packages implemented in R and other programming languages, they either provide only a few optimization algorithms or focus on a specific application field. There does not exist a complete NMF package for the bioinformatics community, and in order to perform various data mining tasks on biological data.

Results

We provide a convenient MATLAB toolbox containing both the implementations of various NMF techniques and a variety of NMF-based data mining approaches for analyzing biological data. Data mining approaches implemented within the toolbox include data clustering and bi-clustering, feature extraction and selection, sample classification, missing values imputation, data visualization, and statistical comparison.

Conclusions

A series of analysis such as molecular pattern discovery, biological process identification, dimension reduction, disease prediction, visualization, and statistical comparison can be performed using this toolbox.

Implications of systemic dysfunction for the etiology of malignancy

The current approach to treatment in oncology is to replace the generally cytotoxic chemotherapies with pharmaceutical treatment which inactivates specific molecular targets associated with cancer development and progression. The goal is to limit cellular damage to pathways perceived to be directly responsible for the malignancy. Its underlying assumptions are twofold: (1) that individual pathways are the cause of malignancy; and (2) that the treatment objective should be destruction-either of the tumor or the dysfunctional pathway. However, the extent to which data actually support these assumptions has not been directly addressed. Accumulating evidence suggests that systemic dysfunction precedes the disruption of specific genetic/molecular pathways in most adult cancers and that targeted treatments such as kinase inhibitors may successfully treat one pathway while generating unintended changes to other, non-targeted pathways. This article discusses (1) the systemic basis of malignancy; (2) better profiling of pre-cancerous biomarkers associated with elevated risk so that preventive lifestyle modifications can be instituted early to revert high-risk epigenetic changes before tumors develop; (3) a treatment emphasis in early stage tumors that would target the restoration of systemic balance by strengthening the body's innate defense mechanisms; and (4) establishing better quantitative models of systems to capture adequate complexity for predictability at all stages of tumor progression.

Identification of aberrant pathways and network activities from high-throughput data

Many complex diseases such as cancer are associated with changes in biological pathways and molecular networks rather than being caused by single gene alterations. A major challenge in the diagnosis and treatment of such diseases is to identify characteristic aberrancies in the biological pathways and molecular network activities and elucidate their relationship to the disease. This review presents recent progress in using high-throughput biological assays to decipher aberrant pathways and network activities. In particular, this review provides specific examples in which high-throughput data have been applied to identify relationships between diseases and aberrant pathways and network activities. The achievements in this field have been remarkable, but many challenges have yet to be addressed.

Gene expression signatures modulated by epidermal growth factor receptor activation and their relationship to cetuximab resistance in head and neck squamous cell carcinoma

Background

Aberrant activation of signaling pathways downstream of epidermal growth factor receptor (EGFR) has been hypothesized to be one of the mechanisms of cetuximab (a monoclonal antibody against EGFR) resistance in head and neck squamous cell carcinoma (HNSCC). To infer relevant and specific pathway activation downstream of EGFR from gene expression in HNSCC, we generated gene expression signatures using immortalized keratinocytes (HaCaT) subjected to ligand stimulation and transfected with EGFR, RELA/p65, or HRAS^Val12D.

Results

The gene expression patterns that distinguished the HaCaT variants and conditions were inferred using the Markov chain Monte Carlo (MCMC) matrix factorization algorithm Coordinated Gene Activity in Pattern Sets (CoGAPS). This approach inferred gene expression signatures with greater relevance to cell signaling pathway activation than the expression signatures inferred with standard linear models. Furthermore, the pathway signature generated using HaCaT-HRAS^Val12D further associated with the cetuximab treatment response in isogenic cetuximab-sensitive (UMSCC1) and -resistant (1CC8) cell lines.

Conclusions

Our data suggest that the CoGAPS algorithm can generate gene expression signatures that are pertinent to downstream effects of receptor signaling pathway activation and potentially be useful in modeling resistance mechanisms to targeted therapies.

Matrix Factorization for Transcriptional Regulatory Network Inference

Inference of Transcriptional Regulatory Networks (TRNs) provides insight into the mechanisms driving biological systems, especially mammalian development and disease. Many techniques have been developed for TRN estimation from indirect biochemical measurements. Although successful when initially tested in model organisms, these regulatory models often fail when applied to data from multicellular organisms where multiple regulation and gene reuse increase dramatically. Non-negative matrix factorization techniques were initially introduced to find non-orthogonal patterns in data, making them ideal techniques for inference in cases of multiple regulation. We review these techniques and their application to TRN analysis.

Array-based pharmacogenomics of molecular-targeted therapies in oncology

The advent of microarrays over the past decade has transformed the way genome-wide studies are designed and conducted, leading to an unprecedented speed of acquisition and amount of new knowledge. Microarray data have led to the identification of molecular subclasses of solid tumors characterized by distinct oncogenic pathways, as well as the development of multigene prognostic or predictive models equivalent or superior to those of established clinical parameters. In the field of molecular-targeted therapy for cancer, in particular, the application of array-based methodologies has enabled the identification of molecular targets with 'key' roles in neoplastic transformation or tumor progression and the subsequent development of targeted agents, which are most likely to be active in the specific molecular setting. Herein, we present a summary of the main applications of whole-genome expression microarrays in the field of molecular-targeted therapies for solid tumors and we discuss their potential in the clinical setting. An emphasis is given on deciphering the molecular mechanisms of drug action, identifying novel therapeutic targets and suitable agents to target them with, and discovering molecular markers/signatures that predict response to therapy or optimal drug dose for each patient.

New perspectives for in vitro risk assessment of multiwalled carbon nanotubes: application of coculture and bioinformatics

Nanotechnology is a rapidly expanding field with wide application for industrial and medical use; therefore, understanding the toxicity of engineered nanomaterials is critical for their commercialization. While short-term in vivo studies have been performed to understand the toxicity profile of various nanomaterials, there is a current effort to shift toxicological testing from in vivo observational models to predictive and high-throughput in vitro models. However, conventional monoculture results of nanoparticle exposure are often disparate and not predictive of in vivo toxic effects. A coculture system of multiple cell types allows for cross-talk between cells and better mimics the in vivo environment. This review proposes that advanced coculture models, combined with integrated analysis of genome-wide in vivo and in vitro toxicogenomic data, may lead to development of predictive multigene expression-based models to better determine toxicity profiles of nanomaterials and consequent potential human health risk due to exposure to these compounds.

Systems Approach to Identifying Relevant Pathways from Phenotype Information in Dose-Dependent Time Series Microarray Data

This study presents a novel computational approach to find relevant pathways from dose-dependent time series gene expression data which are significantly associated with a phenotype pattern pathological patterns in the comprehensive evaluation of database of pathways. Our system uses four steps: 1) identify a set of genes which change significantly in dose or time; 2) find phenotype patterns and gene coefficients for the genes found in step 1; 3) expand to genome-wide coefficients, and 4) identify pathways which are significantly relevant to a phenotype pattern. Our technique finds biologically relevant pathways with and without phenotype-constraints. Our system has been used on genome-wide expression profiles of mouse lungs (n=160) following aspiration of well dispersed multi-walled carbon nanotubes (MWCNT), in order to detect MWCNT-induced lung inflammation and related pathways. The identified significant pathways are supported by evidence in the literature and biological validation.

Chemotherapy and signaling: How can targeted therapies supercharge cytotoxic agents?

In recent years, oncologists have begun to conclude that chemotherapy has reached a plateau of efficacy as a primary treatment modality, even if toxicity can be effectively controlled. Emerging specific inhibitors of signaling and metabolic pathways (i.e., targeted agents) contrast with traditional chemotherapy drugs in that the latter primarily interfere with the DNA biosynthesis and the cell replication machinery. In an attempt to improve on the efficacy, combination of targeted drugs with conventional chemotherapeutics has become a routine way of testing multiple new agents in early phase clinical trials. This review discusses the recent advances including integrative systematic biology and RNAi approaches to counteract the chemotherapy resistance and to buttress the selectivity, efficacy and personalization of anti-cancer drug therapy.

CoGAPS: an R/C++ package to identify patterns and biological process activity in transcriptomic data

SUMMARY

Coordinated Gene Activity in Pattern Sets (CoGAPS) provides an integrated package for isolating gene expression driven by a biological process, enhancing inference of biological processes from transcriptomic data. CoGAPS improves on other enrichment measurement methods by combining a Markov chain Monte Carlo (MCMC) matrix factorization algorithm (GAPS) with a threshold-independent statistic inferring activity on gene sets. The software is provided as open source C++ code built on top of JAGS software with an R interface.

AVAILABILITY

The R package CoGAPS and the C++ package GAPS-JAGS are provided open source under the GNU Lesser Public License (GLPL) with a users manual containing installation and operating instructions. CoGAPS is available through Bioconductor and depends on the rjags package available through CRAN to interface CoGAPS with GAPS-JAGS. URL: http://www.cancerbiostats.onc.jhmi.edu/cogaps.cfm .