Signaling pathways activation profiles make better markers of cancer than expression of individual genes

Transcriptome-Based Traits of Radioresistant Sublines of Non-Small Cell Lung Cancer Cells

Radioresistance is a major obstacle for the successful therapy of many cancers, including non-small cell lung cancer (NSCLC). To elucidate the mechanism of radioresistance of NSCLC cells and to identify key molecules conferring radioresistance, the radioresistant subclones of p53 wild-type A549 and p53-deficient H1299 cell cultures were established. The transcriptional changes between parental and radioresistant NSCLC cells were investigated by RNA-seq. In total, expression levels of 36,596 genes were measured. Changes in the activation of intracellular molecular pathways of cells surviving irradiation relative to parental cells were quantified using the Oncobox bioinformatics platform. Following 30 rounds of 2 Gy irradiation, a total of 322 genes were differentially expressed between p53 wild-type radioresistant A549IR and parental A549 cells. For the p53-deficient (H1299) NSCLC cells, the parental and irradiated populations differed in the expression of 1628 genes and 1616 pathways. The expression of genes associated with radioresistance reflects the complex biological processes involved in clinical cancer cell eradication and might serve as a potential biomarker and therapeutic target for NSCLC treatment.

An improved MOPSO approach with adaptive strategy for identifying biomarkers from gene expression dataset

Biomarkers plays an important role in the prediction and diagnosis of cancers. Therefore, it is urgent to design effective methods to extract biomarkers. The corresponding pathway information of the microarray gene expression data can be obtained from public database, which makes possible to identify biomarkers based on pathway information and has been attracted extensive attention. In the most existing methods, all the member genes in the same pathway are regarded as equally important for inferring pathway activity. However, the contribution of each gene should be different in the process of inferring pathway activity. In this research, an improved multi-objective particle swarm optimization algorithm with penalty boundary intersection decomposition mechanism (IMOPSO-PBI) has been proposed to quantify the relevance of each gene in pathway activity inference. In the proposed algorithm, two optimization objectives namely t-score and z-score respectively has been introduced. In addition, in order to solve the problem that optimal set with poor diversity in the most multi-objective optimization algorithms, an adaptive mechanism for adjusting penalty parameters based on PBI decomposition has been introduced. The performance of the proposed IMOPSO-PBI approach compared with some existing methods on six gene expression datasets has been given. To verify the effectiveness of the proposed IMOPSO-PBI algorithm, experiments were carried out on six gene datasets and the results has been compared with the existing methods. The comparative experiment results show that the proposed IMOPSO-PBI method has a higher classification accuracy and the extracted feature genes are verified possess biological significance.

Personalized targeted therapy prescription in colorectal cancer using algorithmic analysis of RNA sequencing data

BACKGROUND

Overall survival of advanced colorectal cancer (CRC) patients remains poor, and gene expression analysis could potentially complement detection of clinically relevant mutations to personalize CRC treatments.

METHODS

We performed RNA sequencing of formalin-fixed, paraffin-embedded (FFPE) cancer tissue samples of 23 CRC patients and interpreted the data obtained using bioinformatic method Oncobox for expression-based rating of targeted therapeutics. Oncobox ranks cancer drugs according to the efficiency score calculated using target genes expression and molecular pathway activation data. The patients had primary and metastatic CRC with metastases in liver, peritoneum, brain, adrenal gland, lymph nodes and ovary. Two patients had mutations in NRAS, seven others had mutated KRAS gene. Patients were treated by aflibercept, bevacizumab, bortezomib, cabozantinib, cetuximab, crizotinib, denosumab, panitumumab and regorafenib as monotherapy or in combination with chemotherapy, and information on the success of totally 39 lines of therapy was collected.

RESULTS

Oncobox drug efficiency score was effective biomarker that could predict treatment outcomes in the experimental cohort (AUC 0.77 for all lines of therapy and 0.91 for the first line after tumor sampling). Separately for bevacizumab, it was effective in the experimental cohort (AUC 0.87) and in 3 independent literature CRC datasets, n = 107 (AUC 0.84-0.94). It also predicted progression-free survival in univariate (Hazard ratio 0.14) and multivariate (Hazard ratio 0.066) analyses. Difference in AUC scores evidences importance of using recent biosamples for the prediction quality.

CONCLUSION

Our results suggest that RNA sequencing analysis of tumor FFPE materials may be helpful for personalizing prescriptions of targeted therapeutics in CRC.

Transcriptomic Portraits and Molecular Pathway Activation Features of Adult Spinal Intramedullary Astrocytomas

In this study, we report 31 spinal intramedullary astrocytoma (SIA) RNA sequencing (RNA-seq) profiles for 25 adult patients with documented clinical annotations. To our knowledge, this is the first clinically annotated RNA-seq dataset of spinal astrocytomas derived from the intradural intramedullary compartment. We compared these tumor profiles with the previous healthy central nervous system (CNS) RNA-seq data for spinal cord and brain and identified SIA-specific gene sets and molecular pathways. Our findings suggest a trend for SIA-upregulated pathways governing interactions with the immune cells and downregulated pathways for the neuronal functioning in the context of normal CNS activity. In two patient tumor biosamples, we identified diagnostic KIAA1549-BRAF fusion oncogenes, and we also found 16 new SIA-associated fusion transcripts. In addition, we bioinformatically simulated activities of targeted cancer drugs in SIA samples and predicted that several tyrosine kinase inhibitory drugs and thalidomide analogs could be potentially effective as second-line treatment agents to aid in the prevention of SIA recurrence and progression.

EGFR Pathway-Based Gene Signatures of Druggable Gene Mutations in Melanoma, Breast, Lung, and Thyroid Cancers

EGFR, BRAF, PIK3CA, and KRAS genes play major roles in EGFR pathway, and accommodate activating mutations that predict response to many targeted therapeutics. However, connections between these mutations and EGFR pathway expression patterns remain unexplored. Here, we investigated transcriptomic associations with these activating mutations in three ways. First, we compared expressions of these genes in the mutant and wild type tumors, respectively, using RNA sequencing profiles from The Cancer Genome Atlas project database (n = 3660). Second, mutations were associated with the activation level of EGFR pathway. Third, they were associated with the gene signatures of differentially expressed genes from these pathways between the mutant and wild type tumors. We found that the upregulated EGFR pathway was linked with mutations in the BRAF (thyroid cancer, melanoma) and PIK3CA (breast cancer) genes. Gene signatures were associated with BRAF (thyroid cancer, melanoma), EGFR (squamous cell lung cancer), KRAS (colorectal cancer), and PIK3CA (breast cancer) mutations. However, only for the BRAF gene signature in the thyroid cancer we observed strong biomarker diagnostic capacity with AUC > 0.7 (0.809). Next, we validated this signature on the independent literature-based dataset (n = 127, fresh-frozen tissue samples, AUC 0.912), and on the experimental dataset (n = 42, formalin fixed, paraffin embedded tissue samples, AUC 0.822). Our results suggest that the RNA sequencing profiles can be used for robust identification of the replacement of Valine at position 600 with Glutamic acid in the BRAF gene in the papillary subtype of thyroid cancer, and evidence that the specific gene expression levels could provide information about the driver carcinogenic mutations.

Key radioresistance regulation models and marker genes identified by integrated transcriptome analysis in nasopharyngeal carcinoma

Nasopharyngeal carcinoma (NPC) is a malignancy that is endemic to China and Southeast Asia. Radiotherapy is the usual treatment, however, radioresistance remains a major reason for failure. This study aimed to find key radioresistance regulation models and marker genes of NPC and clarify the mechanism of NPC radioresistance by RNA sequencing and bioinformatics analysis of the differences in gene expression profiles between radioresistant and radiosensitive NPC tissues. A total of 21 NPC biopsy specimens with different radiosensitivity were analyzed by RNA sequencing. Differentially expressed genes in RNA sequencing data were identified using R software. The differentially expressed gene data derived from RNA sequencing as well as prior knowledge in the form of pathway databases were integrated to find sub‐networks of related genes. The data of RNA sequencing with the GSE48501 data from the GEO database were combined to further search for more reliable genes associated with radioresistance of NPC. Survival analyses using the Kaplan–Meier method based on the expression of the genes were conducted to facilitate the understanding of the clinical significance of the differentially expressed genes. RT‐qPCR was performed to validate the expression levels of the differentially expressed genes. We identified 1182 differentially expressed genes between radioresistant and radiosensitive NPC tissue samples. Compared to the radiosensitive group, 22 genes were significantly upregulated and 1160 genes were downregulated in the radioresistant group. In addition, 10 major NPC radiation resistance network models were identified through integration analysis with known NPC radiation resistance‐associated genes and mechanisms. Furthermore, we identified three core genes, DOCK4, MCM9, and POPDC3 among 12 common downregulated genes in the two datasets, which were validated by RT‐qPCR. The findings of this study provide new clues for clarifying the mechanism of NPC radioresistance, and further experimental studies of these core genes are warranted.

Machine Learning Applicability for Classification of PAD/VCD Chemotherapy Response Using 53 Multiple Myeloma RNA Sequencing Profiles

Multiple myeloma (MM) affects ~500,000 people and results in ~100,000 deaths annually, being currently considered treatable but incurable. There are several MM chemotherapy treatment regimens, among which eleven include bortezomib, a proteasome-targeted drug. MM patients respond differently to bortezomib, and new prognostic biomarkers are needed to personalize treatments. However, there is a shortage of clinically annotated MM molecular data that could be used to establish novel molecular diagnostics. We report new RNA sequencing profiles for 53 MM patients annotated with responses on two similar chemotherapy regimens: bortezomib, doxorubicin, dexamethasone (PAD), and bortezomib, cyclophosphamide, dexamethasone (VCD), or with responses to their combinations. Fourteen patients received both PAD and VCD; six received only PAD, and 33 received only VCD. We compared profiles for the good and poor responders and found five genes commonly regulated here and in the previous datasets for other bortezomib regimens (all upregulated in the good responders): FGFR3, MAF, IGHA2, IGHV1-69, and GRB14. Four of these genes are linked with known immunoglobulin locus rearrangements. We then used five machine learning (ML) methods to build a classifier distinguishing good and poor responders for two cohorts: PAD + VCD (53 patients), and separately VCD (47 patients). We showed that the application of FloWPS dynamic data trimming was beneficial for all ML methods tested in both cohorts, and also in the previous MM bortezomib datasets. However, the ML models build for the different datasets did not allow cross-transferring, which can be due to different treatment regimens, experimental profiling methods, and MM heterogeneity.

Gene Expression Signature of Endometrial Samples from Women with and without Endometriosis

STUDY OBJECTIVE

To develop a prototype of a complex gene expression biomarker for the diagnosis of endometriosis on the basis of differences between the molecular signatures of the endometrium from women with and without endometriosis.

DESIGN

Prospective observational cohort study. Evidence obtained from a well-designed, controlled trial without randomization.

SETTING

Department of reproductive medicine and surgery, A.I. Evdokimov Moscow State University of Medicine and Dentistry.

PATIENTS

A total of 33 women (aged 32-38 years) were included in this study. Patients with and without endometriosis were divided into 2 separate groups. The group composed of patients with endometriosis included 19 living patients with endometriosis who underwent laparoscopic excision of endometriosis. The control group included 6 living patients who underwent laparoscopic excision of incompetent uterine scar after cesarean section, with both surgically and histologically confirmed absence of endometriosis and adenomyosis. An additional control/verification group included various previously RNA-sequencing-profiled tissue samples (endocervix, ovarian surface epithelium) of 8 randomly selected healthy female cadaveric donors aged 32 to 38 years. The exclusion criteria for all patients were hormone therapy and any intrauterine device use for more than 1 year preceding surgery, as well as absence of other diseases of the uterus, fallopian tubes, and ovaries.

INTERVENTIONS

Laparoscopic excision of endometriotic foci and hysteroscopy with endometrial sampling were performed. The cadaveric tissue samples included endocervix and ovarian surface epithelium. Endometrial sampling was obtained from the women in the control group. RNA sequencing was performed using Illumina HiSeq 3000 equipment (Illumina, Inc., San Diego, CA) for single-end sequencing. Unique bioinformatics algorithms were developed and validated using experimental and public gene expression datasets.

MEASUREMENTS AND MAIN RESULTS

We generated a characteristic signature of 5 genes downregulated in the endometrium and endometriotic tissue of the patients with endometriosis, selected after comparison with the endometrium of the women without endometriosis. This gene signature showed a capacity for nearly perfect separation of all 52 analyzed tissue samples of the patients with endometriosis (endometrial as well as endometriotic samples) from the 14 tissue samples of both living and cadaveric donors without endometriosis (area under the curve = 0.982, Matthews correlation coefficient = 0.832).

CONCLUSION

The gene signature of the endometrium identified in this study may potentially serve as a nonsurgical diagnostic method for endometriosis detection. Our data also suggest that the statistical method of 5-fold cross-validation of differential gene expression analysis can be used to generate robust gene signatures using real-world clinical data.

Using proteomic and transcriptomic data to assess activation of intracellular molecular pathways

Analysis of molecular pathway activation is the recent instrument that helps to quantize activities of various intracellular signaling, structural, DNA synthesis and repair, and biochemical processes. This may have a deep impact in fundamental research, bioindustry, and medicine. Unlike gene ontology analyses and numerous qualitative methods that can establish whether a pathway is affected in principle, the quantitative approach has the advantage of exactly measuring the extent of a pathway up/downregulation. This results in emergence of a new generation of molecular biomarkers-pathway activation levels, which reflect concentration changes of all measurable pathway components. The input data can be the high-throughput proteomic or transcriptomic profiles, and the output numbers take both positive and negative values and positively reflect overall pathway activation. Due to their nature, the pathway activation levels are more robust biomarkers compared to the individual gene products/protein levels. Here, we review the current knowledge of the quantitative gene expression interrogation methods and their applications for the molecular pathway quantization. We consider enclosed bioinformatic algorithms and their applications for solving real-world problems. Besides a plethora of applications in basic life sciences, the quantitative pathway analysis can improve molecular design and clinical investigations in pharmaceutical industry, can help finding new active biotechnological components and can significantly contribute to the progressive evolution of personalized medicine. In addition to the theoretical principles and concepts, we also propose publicly available software for the use of large-scale protein/RNA expression data to assess the human pathway activation levels.

Algorithmic Annotation of Functional Roles for Components of 3,044 Human Molecular Pathways

Current methods of high-throughput molecular and genomic analyses enabled to reconstruct thousands of human molecular pathways. Knowledge of molecular pathways structure and architecture taken along with the gene expression data can help interrogating the pathway activation levels (PALs) using different bioinformatic algorithms. In turn, the pathway activation profiles can characterize molecular processes, which are differentially regulated and give numeric characteristics of the extent of their activation or inhibition. However, different pathway nodes may have different functions toward overall pathway regulation, and calculation of PAL requires knowledge of molecular function of every node in the pathway in terms of its activator or inhibitory role. Thus, high-throughput annotation of functional roles of pathway nodes is required for the comprehensive analysis of the pathway activation profiles. We proposed an algorithm that identifies functional roles of the pathway components and applied it to annotate 3,044 human molecular pathways extracted from the Biocarta, Reactome, KEGG, Qiagen Pathway Central, NCI, and HumanCYC databases and including 9,022 gene products. The resulting knowledgebase can be applied for the direct calculation of the PALs and establishing large scale profiles of the signaling, metabolic, and DNA repair pathway regulation using high throughput gene expression data. We also provide a bioinformatic tool for PAL data calculations using the current pathway knowledgebase.

System, Method and Software for Calculation of a Cannabis Drug Efficiency Index for the Reduction of Inflammation

There are many varieties of Cannabis sativa that differ from each other by composition of cannabinoids, terpenes and other molecules. The medicinal properties of these cultivars are often very different, with some being more efficient than others. This report describes the development of a method and software for the analysis of the efficiency of various cannabis extracts to detect the anti-inflammatory properties of the various cannabis extracts. The method uses high-throughput gene expression profiling data but can potentially use other omics data as well. According to the signaling pathway topology, the gene expression profiles are convoluted into the signaling pathway activities using a signaling pathway impact analysis (SPIA) method. The method was tested by inducing inflammation in human 3D epithelial tissues, including intestine, oral and skin, and then exposing these tissues to various extracts and then performing transcriptome analysis. The analysis showed a different efficiency of the various extracts in restoring the transcriptome changes to the pre-inflammation state, thus allowing to calculate a different cannabis drug efficiency index (CDEI).

Cancer gene expression profiles associated with clinical outcomes to chemotherapy treatments

Background

Machine learning (ML) methods still have limited applicability in personalized oncology due to low numbers of available clinically annotated molecular profiles. This doesn’t allow sufficient training of ML classifiers that could be used for improving molecular diagnostics.

Methods

We reviewed published datasets of high throughput gene expression profiles corresponding to cancer patients with known responses on chemotherapy treatments. We browsed Gene Expression Omnibus (GEO), The Cancer Genome Atlas (TCGA) and Tumor Alterations Relevant for GEnomics-driven Therapy (TARGET) repositories.

Results

We identified data collections suitable to build ML models for predicting responses on certain chemotherapeutic schemes. We identified 26 datasets, ranging from 41 till 508 cases per dataset. All the datasets identified were checked for ML applicability and robustness with leave-one-out cross validation. Twenty-three datasets were found suitable for using ML that had balanced numbers of treatment responder and non-responder cases.

Conclusions

We collected a database of gene expression profiles associated with clinical responses on chemotherapy for 2786 individual cancer cases. Among them seven datasets included RNA sequencing data (for 645 cases) and the others – microarray expression profiles. The cases represented breast cancer, lung cancer, low-grade glioma, endothelial carcinoma, multiple myeloma, adult leukemia, pediatric leukemia and kidney tumors. Chemotherapeutics included taxanes, bortezomib, vincristine, trastuzumab, letrozole, tipifarnib, temozolomide, busulfan and cyclophosphamide.

RNA Sequencing in Comparison to Immunohistochemistry for Measuring Cancer Biomarkers in Breast Cancer and Lung Cancer Specimens

RNA sequencing is considered the gold standard for high-throughput profiling of gene expression at the transcriptional level. Its increasing importance in cancer research and molecular diagnostics is reflected in the growing number of its mentions in scientific literature and clinical trial reports. However, the use of different reagents and protocols for RNA sequencing often produces incompatible results. Recently, we published the Oncobox Atlas of RNA sequencing profiles for normal human tissues obtained from healthy donors killed in road accidents. This is a database of molecular profiles obtained using uniform protocol and reagents settings that can be broadly used in biomedicine for data normalization in pathology, including cancer. Here, we publish new original 39 breast cancer (BC) and 19 lung cancer (LC) RNA sequencing profiles obtained for formalin-fixed paraffin-embedded (FFPE) tissue samples, fully compatible with the Oncobox Atlas. We performed the first correlation study of RNA sequencing and immunohistochemistry-measured expression profiles for the clinically actionable biomarker genes in FFPE cancer tissue samples. We demonstrated high (Spearman's rho 0.65-0.798) and statistically significant (p < 0.00004) correlations between the RNA sequencing (Oncobox protocol) and immunohistochemical measurements for HER2/ERBB2, ER/ESR1 and PGR genes in BC, and for PDL1 gene in LC; AUC: 0.963 for HER2, 0.921 for ESR1, 0.912 for PGR, and 0.922 for PDL1. To our knowledge, this is the first validation that total RNA sequencing of archived FFPE materials provides a reliable estimation of marker protein levels. These results show that in the future, RNA sequencing can complement immunohistochemistry for reliable measurements of the expression biomarkers in FFPE cancer samples.

RNA sequencing profiles and diagnostic signatures linked with response to ramucirumab in gastric cancer

Gastric cancer (GC) is the fifth-ranked cancer type by associated mortality. The proportion of early diagnosis is low, and most patients are diagnosed at the advanced stages. First-line therapy standardly includes fluoropyrimidines and platinum compounds with trastuzumab for HER2-positive cases. For recurrent disease, there are several alternative options including ramucirumab, a monoclonal therapeutic antibody that inhibits VEGF-mediated tumor angiogenesis by binding with VEGFR2, alone or in combination with other cancer drugs. However, overall response rate following ramucirumab or its combinations is 30%-80% of the patients, suggesting that personalization of drug prescription is needed to increase efficacy of treatment. We report here original tumor RNA sequencing profiles for 15 advanced GC patients linked with data on clinical response to ramucirumab or its combinations. Three genes showed differential expression in the tumors for responders versus nonresponders: CHRM3, LRFN1, and TEX15 Of them, CHRM3 was up-regulated in the responders. Using the bioinformatic platform Oncobox we simulated ramucirumab efficiency and compared output model results with actual tumor response data. An agreement was observed between predicted and real clinical outcomes (AUC ≥ 0.7). These results suggest that RNA sequencing may be used to personalize the prescription of ramucirumab for GC and indicate potential molecular mechanisms underlying ramucirumab resistance. The RNA sequencing profiles obtained here are fully compatible with the previously published Oncobox Atlas of Normal Tissue Expression (ANTE) data.

Gene signatures associated with genomic aberrations predict prognosis in neuroblastoma

BACKGROUND

Neuroblastoma (NB) is a heterogeneous disease with respect to genomic abnormalities and clinical behaviors. Despite recent advances in our understanding of the association between the genetic aberrations and clinical features, it remains one of the major challenges to predict prognosis and stratify patients for determining personalized therapy in this disease. The aim of this study was to develop an effective prognosis prediction model for NB patients.

METHODS

We integrated diverse computational analyses to define gene signatures that reflect MYCN activity and chromosomal aberrations including deletion of chromosome 1p (Chr1p_del) and chromosome 11q (Chr11q_del) as well as chromosome 11q whole loss (Chr11q_wls). We evaluated the prognostic and predictive values of these signatures in seven NB gene expression datasets (the number of samples ranges from 94 to 498, with a total of 2120) generated from both RNA sequencing and microarray platforms.

RESULTS

MYCN signature was a more effective prognostic marker than MYCN amplification status and MYCN expression. Similarly, the Chr1p_del score was more prognostic than Chr1p status. The activity scores of MYCN, Chr1p_del and Chr11q_del were associated with poor prognosis, while the Chr11q_wls score was linked to good outcome. We integrated the activity scores of MYCN, Chr1p_del, Chr11q_del, and Chr11q_wls and clinical variables into an integrative prognostic model, which displayed significant performance over the clinical variables or each genomic aberration alone.

CONCLUSIONS

Our integrative gene signature model shows a significantly improved forecast performance with prognostic and predictive information, and thereby can be served as a biomarker to stratify NB patients for prognosis evaluation and surveillance programs.

Intratumoral Heterogeneity and Longitudinal Changes in Gene Expression Predict Differential Drug Sensitivity in Newly Diagnosed and Recurrent Glioblastoma

Background: Inevitable recurrence after radiochemotherapy is the major problem in the treatment of glioblastoma, the most prevalent type of adult brain malignancy. Glioblastomas are notorious for a high degree of intratumor heterogeneity manifest through a diversity of cell types and molecular patterns. The current paradigm of understanding glioblastoma recurrence is that cytotoxic therapy fails to target effectively glioma stem cells. Recent advances indicate that therapy-driven molecular evolution is a fundamental trait associated with glioblastoma recurrence. There is a growing body of evidence indicating that intratumor heterogeneity, longitudinal changes in molecular biomarkers and specific impacts of glioma stem cells need to be taken into consideration in order to increase the accuracy of molecular diagnostics still relying on readouts obtained from a single tumor specimen. Methods: This study integrates a multisampling strategy, longitudinal approach and complementary transcriptomic investigations in order to identify transcriptomic traits of recurrent glioblastoma in whole-tissue specimens of glioblastoma or glioblastoma stem cells. In this study, 128 tissue samples of 44 tumors including 23 first diagnosed, 19 recurrent and 2 secondary recurrent glioblastomas were analyzed along with 27 primary cultures of glioblastoma stem cells by RNA sequencing. A novel algorithm was used to quantify longitudinal changes in pathway activities and model efficacy of anti-cancer drugs based on gene expression data. Results: Our study reveals that intratumor heterogeneity of gene expression patterns is a fundamental characteristic of not only newly diagnosed but also recurrent glioblastomas. Evidence is provided that glioblastoma stem cells recapitulate intratumor heterogeneity, longitudinal transcriptomic changes and drug sensitivity patterns associated with the state of recurrence. Conclusions: Our results provide a transcriptional rationale for the lack of significant therapeutic benefit from temozolomide in patients with recurrent glioblastoma. Our findings imply that the spectrum of potentially effective drugs is likely to differ between newly diagnosed and recurrent glioblastomas and underscore the merits of glioblastoma stem cells as prognostic models for identifying alternative drugs and predicting drug response in recurrent glioblastoma. With the majority of recurrent glioblastomas being inoperable, glioblastoma stem cell models provide the means of compensating for the limited availability of recurrent glioblastoma specimens.

Mutation Enrichment and Transcriptomic Activation Signatures of 419 Molecular Pathways in Cancer

Carcinogenesis is linked with massive changes in regulation of gene networks. We used high throughput mutation and gene expression data to interrogate involvement of 278 signaling, 72 metabolic, 48 DNA repair and 47 cytoskeleton molecular pathways in cancer. Totally, we analyzed 4910 primary tumor samples with individual cancer RNA sequencing and whole exome sequencing profiles including ~1.3 million DNA mutations and representing thirteen cancer types. Gene expression in cancers was compared with the corresponding 655 normal tissue profiles. For the first time, we calculated mutation enrichment values and activation levels for these pathways. We found that pathway activation profiles were largely congruent among the different cancer types. However, we observed no correlation between mutation enrichment and expression changes both at the gene and at the pathway levels. Overall, positive median cancer-specific activation levels were seen in the DNA repair, versus similar slightly negative values in the other types of pathways. The DNA repair pathways also demonstrated the highest values of mutation enrichment. However, the signaling and cytoskeleton pathways had the biggest proportions of representatives among the outstandingly frequently mutated genes thus suggesting their initiator roles in carcinogenesis and the auxiliary/supporting roles for the other groups of molecular pathways.

Molecular Pathway Analysis of Mutation Data for Biomarkers Discovery and Scoring of Target Cancer Drugs

DNA mutations govern cancer development. Cancer mutation profiles vary dramatically among the individuals. In some cases, they may serve as the predictors of disease progression and response to therapies. However, the biomarker potential of cancer mutations can be dramatically (several orders of magnitude) enhanced by applying molecular pathway-based approach. We developed Oncobox system for calculation of pathway instability (PI) values for the molecular pathways that are aggregated mutation frequencies of the pathway members normalized on gene lengths and on number of genes in the pathway. PI scores can be effective biomarkers in different types of comparisons, for example, as the cancer type biomarkers and as the predictors of tumor response to target therapies. The latter option is implemented using mutation drug score (MDS) values, which algorithmically rank the drugs capacity of interfering with the mutated molecular pathways. Here, describe the mathematical basis and algorithms for PI and MDS values calculation, validation and implementation. The example analysis is provided encompassing 5956 human tumor mutation profiles of 15 cancer types from The Cancer Genome Atlas (TCGA) project, that totally make 2,316,670 mutations in 19,872 genes and 1748 molecular pathways, thus enabling ranking of 128 clinically approved target drugs. Our results evidence that the Oncobox PI and MDS approaches are highly useful for basic and applied aspects of molecular oncology and pharmacology research.

H3K4me3, H3K9ac, H3K27ac, H3K27me3 and H3K9me3 Histone Tags Suggest Distinct Regulatory Evolution of Open and Condensed Chromatin Landmarks

BACKGROUND

Transposons are selfish genetic elements that self-reproduce in host DNA. They were active during evolutionary history and now occupy almost half of mammalian genomes. Close insertions of transposons reshaped structure and regulation of many genes considerably. Co-evolution of transposons and host DNA frequently results in the formation of new regulatory regions. Previously we published a concept that the proportion of functional features held by transposons positively correlates with the rate of regulatory evolution of the respective genes.

METHODS

We ranked human genes and molecular pathways according to their regulatory evolution rates based on high throughput genome-wide data on five histone modifications (H3K4me3, H3K9ac, H3K27ac, H3K27me3, H3K9me3) linked with transposons for five human cell lines.

RESULTS

Based on the total of approximately 1.5 million histone tags, we ranked regulatory evolution rates for 25075 human genes and 3121 molecular pathways and identified groups of molecular processes that showed signs of either fast or slow regulatory evolution. However, histone tags showed different regulatory patterns and formed two distinct clusters: promoter/active chromatin tags (H3K4me3, H3K9ac, H3K27ac) vs. heterochromatin tags (H3K27me3, H3K9me3).

CONCLUSION

In humans, transposon-linked histone marks evolved in a coordinated way depending on their functional roles.

RNA sequencing for research and diagnostics in clinical oncology

Molecular diagnostics is becoming one of the major drivers of personalized oncology. With hundreds of different approved anticancer drugs and regimens of their administration, selecting the proper treatment for a patient is at least nontrivial task. This is especially sound for the cases of recurrent and metastatic cancers where the standard lines of therapy failed. Recent trials demonstrated that mutation assays have a strong limitation in personalized selection of therapeutics, consequently, most of the drugs cannot be ranked and only a small percentage of patients can benefit from the screening. Other approaches are, therefore, needed to address a problem of finding proper targeted therapies. The analysis of RNA expression (transcriptomic) profiles presents a reasonable solution because transcriptomics stands a few steps closer to tumor phenotype than the genome analysis. Several recent studies pioneered using transcriptomics for practical oncology and showed truly encouraging clinical results. The possibility of directly measuring of expression levels of molecular drugs' targets and profiling activation of the relevant molecular pathways enables personalized prioritizing for all types of molecular-targeted therapies. RNA sequencing is the most robust tool for the high throughput quantitative transcriptomics. Its use, potentials, and limitations for the clinical oncology will be reviewed here along with the technical aspects such as optimal types of biosamples, RNA sequencing profile normalization, quality controls and several levels of data analysis.

Retroelement-Linked Transcription Factor Binding Patterns Point to Quickly Developing Molecular Pathways in Human Evolution

Background: Retroelements (REs) are transposable elements occupying ~40% of the human genome that can regulate genes by providing transcription factor binding sites (TFBS). RE-linked TFBS profile can serve as a marker of gene transcriptional regulation evolution. This approach allows for interrogating the regulatory evolution of organisms with RE-rich genomes. We aimed to characterize the evolution of transcriptional regulation for human genes and molecular pathways using RE-linked TFBS accumulation as a metric. Methods: We characterized human genes and molecular pathways either enriched or deficient in RE-linked TFBS regulation. We used ENCODE database with mapped TFBS for 563 transcription factors in 13 human cell lines. For 24,389 genes and 3124 molecular pathways, we calculated the score of RE-linked TFBS regulation reflecting the regulatory evolution rate at the level of individual genes and molecular pathways. Results: The major groups enriched by RE regulation deal with gene regulation by microRNAs, olfaction, color vision, fertilization, cellular immune response, and amino acids and fatty acids metabolism and detoxication. The deficient groups were involved in translation, RNA transcription and processing, chromatin organization, and molecular signaling. Conclusion: We identified genes and molecular processes that have characteristics of especially high or low evolutionary rates at the level of RE-linked TFBS regulation in human lineage.

Pathway Based Analysis of Mutation Data Is Efficient for Scoring Target Cancer Drugs

Despite the significant achievements in chemotherapy, cancer remains one of the leading causes of death. Target therapy revolutionized this field, but efficiencies of target drugs show dramatic variation among individual patients. Personalization of target therapies remains, therefore, a challenge in oncology. Here, we proposed molecular pathway-based algorithm for scoring of target drugs using high throughput mutation data to personalize their clinical efficacies. This algorithm was validated on 3,800 exome mutation profiles from The Cancer Genome Atlas (TCGA) project for 128 target drugs. The output values termed Mutational Drug Scores (MDS) showed positive correlation with the published drug efficiencies in clinical trials. We also used MDS approach to simulate all known protein coding genes as the putative drug targets. The model used was built on the basis of 18,273 mutation profiles from COSMIC database for eight cancer types. We found that the MDS algorithm-predicted hits frequently coincide with those already used as targets of the existing cancer drugs, but several novel candidates can be considered promising for further developments. Our results evidence that the MDS is applicable to ranking of anticancer drugs and can be applied for the identification of novel molecular targets.

Pathway Instability Is an Effective New Mutation-Based Type of Cancer Biomarkers

DNA mutations play a crucial role in cancer development and progression. Mutation profiles vary dramatically in different cancer types and between individual tumors. Mutations of several individual genes are known as reliable cancer biomarkers, although the number of such genes is tiny and does not enable differential diagnostics for most of the cancers. We report here a technique enabling dramatically increased efficiency of cancer biomarkers development using DNA mutations data. It includes a quantitative metric termed Pathway instability (PI) based on mutations enrichment of intracellular molecular pathways. This method was tested on 5,956 tumor mutation profiles of 15 cancer types from The Cancer Genome Atlas (TCGA) project. Totally, we screened 2,316,670 mutations in 19,872 genes and 1,748 molecular pathways. Our results demonstrated considerable advantage of pathway-based mutation biomarkers over individual gene mutation profiles, as reflected by more than two orders of magnitude greater numbers by high-quality [ROC area-under-curve (AUC)>0.75] biomarkers. For example, the number of such high-quality mutational biomarkers distinguishing between different cancer types was only six for the individual gene mutations, and already 660 for the pathway-based biomarkers. These results evidence that PI value can be used as a new generation of complex cancer biomarkers significantly outperforming the existing gene mutation biomarkers.

Integrated transcriptomic and epigenomic analysis of ovarian cancer reveals epigenetically silenced GULP1

Many epigenetically inactivated genes involved in ovarian cancer (OC) development and progression remain to be identified. In this study we undertook an integrated approach that consisted of identification of genome-wide expression patterns of primary OC samples and normal ovarian surface epithelium along with a pharmacologic unmasking strategy using 3 OC and 3 immortalized normal ovarian epithelial cell lines. Our filtering scheme identified 43 OC specific methylated genes and among the 5 top candidates (GULP1, CLIP4, BAMBI, NT5E, TGFβ2), we performed extended studies of GULP1. In a training set, we identified GULP1 methylation in 21/61 (34%) of cases with 100% specificity. In an independent cohort, the observed methylation was 40% (146/365) in OC, 12.5% (2/16) in borderline tumors, 11% (2/18) in cystadenoma and 0% (0/13) in normal ovarian epithelium samples. GULP1 methylation was associated with clinicopathological parameters such as stage III/IV (p = 0.001), poorly differentiated grade (p = 0.033), residual disease (p < 0.0003), worse overall (p = 0.02) and disease specific survival (p = 0.01). Depletion of GULP1 in OC cells led to increased pro-survival signaling, inducing survival and colony formation, whereas reconstitution of GULP1 negated these effects, suggesting that GULP1 is required for maintaining cellular growth control.

Large-scale profiling of signalling pathways reveals an asthma specific signature in bronchial smooth muscle cells

Background

Bronchial smooth muscle (BSM) cells from asthmatic patients maintain in vitro a distinct hyper-reactive (“primed”) phenotype, characterized by increased release of pro-inflammatory factors and mediators, as well as hyperplasia and/or hypertrophy. This “primed” phenotype helps to understand pathogenesis of asthma, as changes in BSM function are essential for manifestation of allergic and inflammatory responses and airway wall remodelling.

Objective

To identify signalling pathways in cultured primary BSMs of asthma patients and non-asthmatic subjects by genome wide profiling of differentially expressed mRNAs and activated intracellular signalling pathways (ISPs).

Methods

Transcriptome profiling by cap-analysis-of-gene-expression (CAGE), which permits selection of preferentially capped mRNAs most likely to be translated into proteins, was performed in human BSM cells from asthmatic (n=8) and non-asthmatic (n=6) subjects and OncoFinder tool were then exploited for identification of ISP deregulations.

Results

CAGE revealed >600 RNAs differentially expressed in asthma vs control cells (p≤0.005), with asthma samples showing a high degree of similarity among them. Comprehensive ISP activation analysis revealed that among 269 pathways analysed, 145 (p<0.05) or 103 (p<0.01) are differentially active in asthma, with profiles that clearly characterize BSM cells of asthmatic individuals. Notably, we identified 7 clusters of coherently acting pathways functionally related to the disease, with ISPs down-regulated in asthma mostly targeting cell death-promoting pathways and up-regulated ones affecting cell growth and proliferation, inflammatory response, control of smooth muscle contraction and hypoxia-related signalization.

Conclusions

These first-time results can now be exploited toward development of novel therapeutic strategies targeting ISP signatures linked to asthma pathophysiology.

Use of deep neural network ensembles to identify embryonic-fetal transition markers: repression of COX7A1 in embryonic and cancer cells

Here we present the application of deep neural network (DNN) ensembles trained on transcriptomic data to identify the novel markers associated with the mammalian embryonic-fetal transition (EFT). Molecular markers of this process could provide important insights into regulatory mechanisms of normal development, epimorphic tissue regeneration and cancer. Subsequent analysis of the most significant genes behind the DNNs classifier on an independent dataset of adult-derived and human embryonic stem cell (hESC)-derived progenitor cell lines led to the identification of COX7A1 gene as a potential EFT marker. COX7A1, encoding a cytochrome C oxidase subunit, was up-regulated in post-EFT murine and human cells including adult stem cells, but was not expressed in pre-EFT pluripotent embryonic stem cells or their in vitro-derived progeny. COX7A1 expression level was observed to be undetectable or low in multiple sarcoma and carcinoma cell lines as compared to normal controls. The knockout of the gene in mice led to a marked glycolytic shift reminiscent of the Warburg effect that occurs in cancer cells. The DNN approach facilitated the elucidation of a potentially new biomarker of cancer and pre-EFT cells, the embryo-onco phenotype, which may potentially be used as a target for controlling the embryonic-fetal transition.

In search for geroprotectors: in silico screening and in vitro validation of signalome-level mimetics of young healthy state

Populations in developed nations throughout the world are rapidly aging, and the search for geroprotectors, or anti-aging interventions, has never been more important. Yet while hundreds of geroprotectors have extended lifespan in animal models, none have yet been approved for widespread use in humans. GeroScope is a computational tool that can aid prediction of novel geroprotectors from existing human gene expression data. GeroScope maps expression differences between samples from young and old subjects to aging-related signaling pathways, then profiles pathway activation strength (PAS) for each condition. Known substances are then screened and ranked for those most likely to target differential pathways and mimic the young signalome. Here we used GeroScope and shortlisted ten substances, all of which have lifespan-extending effects in animal models, and tested 6 of them for geroprotective effects in senescent human fibroblast cultures. PD-98059, a highly selective MEK1 inhibitor, showed both life-prolonging and rejuvenating effects. Natural compounds like N-acetyl-L-cysteine, Myricetin and Epigallocatechin gallate also improved several senescence-associated properties and were further investigated with pathway analysis. This work not only highlights several potential geroprotectors for further study, but also serves as a proof-of-concept for GeroScope, Oncofinder and other PAS-based methods in streamlining drug prediction, repurposing and personalized medicine.

Towards natural mimetics of metformin and rapamycin

Aging is now at the forefront of major challenges faced globally, creating an immediate need for safe, widescale interventions to reduce the burden of chronic disease and extend human healthspan. Metformin and rapamycin are two FDA-approved mTOR inhibitors proposed for this purpose, exhibiting significant anti-cancer and anti-aging properties beyond their current clinical applications. However, each faces issues with approval for off-label, prophylactic use due to adverse effects. Here, we initiate an effort to identify nutraceuticals-safer, naturally-occurring compounds-that mimic the anti-aging effects of metformin and rapamycin without adverse effects. We applied several bioinformatic approaches and deep learning methods to the Library of Integrated Network-based Cellular Signatures (LINCS) dataset to map the gene- and pathway-level signatures of metformin and rapamycin and screen for matches among over 800 natural compounds. We then predicted the safety of each compound with an ensemble of deep neural network classifiers. The analysis revealed many novel candidate metformin and rapamycin mimetics, including allantoin and ginsenoside (metformin), epigallocatechin gallate and isoliquiritigenin (rapamycin), and withaferin A (both). Four relatively unexplored compounds also scored well with rapamycin. This work revealed promising candidates for future experimental validation while demonstrating the applications of powerful screening methods for this and similar endeavors.

Data aggregation at the level of molecular pathways improves stability of experimental transcriptomic and proteomic data

High throughput technologies opened a new era in biomedicine by enabling massive analysis of gene expression at both RNA and protein levels. Unfortunately, expression data obtained in different experiments are often poorly compatible, even for the same biologic samples. Here, using experimental and bioinformatic investigation of major experimental platforms, we show that aggregation of gene expression data at the level of molecular pathways helps to diminish cross- and intra-platform bias otherwise clearly seen at the level of individual genes. We created a mathematical model of cumulative suppression of data variation that predicts the ideal parameters and the optimal size of a molecular pathway. We compared the abilities to aggregate experimental molecular data for the 5 alternative methods, also evaluated by their capacity to retain meaningful features of biologic samples. The bioinformatic method OncoFinder showed optimal performance in both tests and should be very useful for future cross-platform data analyses.

Gene expression and molecular pathway activation signatures of MYCN-amplified neuroblastomas

Neuroblastoma is a pediatric cancer arising from sympathetic nervous system. Remarkable heterogeneity in outcomes is one of its widely known features. One of the traits strongly associated with the unfavorable subtype is the amplification of oncogene MYCN. Here, we performed cross-platform biomarker detection by comparing gene expression and pathway activation patterns from the two literature reports and from our experimental dataset, combining profiles for the 761 neuroblastoma patients with known MYCN amplification status. We identified 109 / 25 gene expression / pathway activation biomarkers strongly linked with the MYCN amplification. The marker genes/pathways are involved in the processes of purine nucleotide biosynthesis, ATP-binding, tetrahydrofolate metabolism, building mitochondrial matrix, biosynthesis of amino acids, tRNA aminoacylation and NADP-linked oxidation-reduction processes, as well as in the tyrosine phosphatase activity, p53 signaling, cell cycle progression and the G1/S and G2/M checkpoints. To connect molecular functions of the genes involved in MYCN-amplified phenotype, we built a new molecular pathway using known intracellular protein interaction networks. The activation of this pathway was highly selective in discriminating MYCN-amplified neuroblastomas in all three datasets. Our data also suggest that the phosphoinositide 3-kinase (PI3K) inhibitors may provide new opportunities for the treatment of the MYCN-amplified neuroblastoma subtype.

Differentially expressed proteins in glioblastoma multiforme identified with a nanobody-based anti-proteome approach and confirmed by OncoFinder as possible tumor-class predictive biomarker candidates

Glioblastoma multiforme is the most frequent primary malignancy of the central nervous system. Despite remarkable progress towards an understanding of tumor biology, there is no efficient treatment and patient outcome remains poor. Here, we present a unique anti-proteomic approach for selection of nanobodies specific for overexpressed glioblastoma proteins. A phage-displayed nanobody library was enriched in protein extracts from NCH644 and NCH421K glioblastoma cell lines. Differential ELISA screenings revealed seven nanobodies that target the following antigens: the ACTB/NUCL complex, VIM, NAP1L1, TUFM, DPYSL2, CRMP1, and ALYREF. Western blots showed highest protein up-regulation for ALYREF, CRMP1, and VIM. Moreover, bioinformatic analysis with the OncoFinder software against the complete "Cancer Genome Atlas" brain tumor gene expression dataset suggests the involvement of different proteins in the WNT and ATM pathways, and in Aurora B, Sem3A, and E-cadherin signaling. We demonstrate the potential use of NAP1L1, NUCL, CRMP1, ACTB, and VIM for differentiation between glioblastoma and lower grade gliomas, with DPYSL2 as a promising "glioma versus reference" biomarker. A small scale validation study confirmed significant changes in mRNA expression levels of VIM, DPYSL2, ACTB and TRIM28. This work helps to fill the information gap in this field by defining novel differences in biochemical profiles between gliomas and reference samples. Thus, selected genes can be used to distinguish glioblastoma from lower grade gliomas, and from reference samples. These findings should be valuable for glioblastoma patients once they are validated on a larger sample size.

Chemo brain: From discerning mechanisms to lifting the brain fog-An aging connection

Mounting evidence indicates that cancer treatments cause numerous deleterious effects, including central nervous system (CNS) toxicity. Chemotherapy-caused CNS side effects encompass changes in cognitive function, memory, and attention, to name a few. Although chemotherapy treatment-induced side effects occur in 16-75% of all patients, the mechanisms of these effects are not well understood. We have recently proposed a new epigenetic theory of chemo brain and, in a pioneer study, determined that cytotoxic chemotherapy agents induce oxidative DNA damage and affect molecular and epigenetic processes in the brain, and may be associated with brain aging processes. In this paper, we discuss the implications of chemo brain epigenetic effects and future perspectives, as well as outline potential links with brain aging and future translational research opportunities.

An analysis of gene expression data involving examination of signaling pathways activation reveals new insights into the mechanism of action of minoxidil topical foam in men with androgenetic alopecia

Androgenetic alopecia is the most common form of hair loss. Minoxidil has been approved for the treatment of hair loss, however its mechanism of action is still not fully clarified. In this study, we aimed to elucidate the effects of 5% minoxidil topical foam on gene expression and activation of signaling pathways in vertex and frontal scalp of men with androgenetic alopecia. We identified regional variations in gene expression and perturbed signaling pathways using in silico Pathway Activation Network Decomposition Analysis (iPANDA) before and after treatment with minoxidil. Vertex and frontal scalp of patients showed a generally similar response to minoxidil. Both scalp regions showed upregulation of genes that encode keratin associated proteins and downregulation of ILK, Akt, and MAPK signaling pathways after minoxidil treatment. Our results provide new insights into the mechanism of action of minoxidil topical foam in men with androgenetic alopecia.

Biomarkers in hepatocellular carcinoma: an overview

Current methods for HCC diagnosis have not an optimal diagnostic accuracy. The detection of more than one biomarker seems to improve their individual performance and provide an accurate HCC diagnosis approach. Individual gene expression seems to influence whether or not the treatment is successful, since several molecules have interfere with cancer associated pathways and have been related to poor prognosis which condition the lack of effective treatment options. Areas covered: Novel biomarkers have been proposed as a useful tool in each patient prognosis. This article aims to review the recent evidence based on HCC biomarkers which seems to have a regulative role according to tumor cell development leading to a specific biological response. Epigenetic regulation, miRNAs, and genome sequencing analysis propose molecular expression signatures as novel biomarkers which allowed achieve the major goal for the use of biomarkers in clinical practice. Moreover, a deeper analysis for determine the diagnostic accuracy of biomarkers has been made. Expert commentary: To improve of methodological designs and sample sizes are needed in order to support the role of biomarkers in HCC. Furthermore, is necessary to consider HCC etiologies and all clinic disease context to carried out clinical phase studies to thrust biomarkers application.

In silico analysis of pathways activation landscape in oral squamous cell carcinoma and oral leukoplakia

A subset of patients with oral squamous cell carcinoma (OSCC), the most common subtype of head and neck squamous cell carcinoma (HNSCC), harbor dysplastic lesions (often visually identified as leukoplakia) prior to cancer diagnosis. Although evidence suggest that leukoplakia represents an initial step in the progression to cancer, signaling networks driving this progression are poorly understood. Here, we applied in silico Pathway Activation Network Decomposition Analysis (iPANDA), a new bioinformatics software suite for qualitative analysis of intracellular signaling pathway activation using transcriptomic data, to assess a network of molecular signaling in OSCC and pre-neoplastic oral lesions. In tumor samples, our analysis detected major conserved mitogenic and survival signaling pathways strongly associated with HNSCC, suggesting that some of the pathways identified by our algorithm, but not yet validated as HNSCC related, may be attractive targets for future research. While pathways activation landscape in the majority of leukoplakias was different from that seen in OSCC, a subset of pre-neoplastic lesions has demonstrated some degree of similarity to the signaling profile seen in tumors, including dysregulation of the cancer-driving pathways related to survival and apoptosis. These results suggest that dysregulation of these signaling networks may be the driving force behind the early stages of OSCC tumorigenesis. While future studies with larger leukoplakia data sets are warranted to further estimate the values of this approach for capturing signaling features that characterize relevant lesions that actually progress to cancers, our platform proposes a promising new approach for detecting cancer-promoting pathways and tailoring the right therapy to prevent tumorigenesis.

Pro-fibrotic pathway activation in trabecular meshwork and lamina cribrosa is the main driving force of glaucoma

While primary open-angle glaucoma (POAG) is a leading cause of blindness worldwide, it still does not have a clear mechanism that can explain all clinical cases of the disease. Elevated IOP is associated with increased accumulation of extracellular matrix (ECM) proteins in the trabecular meshwork (TM) that prevents normal outflow of aqueous humor (AH) and has damaging effects on the fine mesh-like lamina cribrosa (LC) through which the optic nerve fibers pass. Applying a pathway analysis algorithm, we discovered that an elevated level of TGFβ observed in glaucoma-affected tissues could lead to pro-fibrotic pathway activation in TM and in LC. In turn, activated pro-fibrotic pathways lead to ECM remodeling in TM and LC, making TM less efficient in AH drainage and making LC more susceptible to damage from elevated IOP via ECM transformation in LC. We propose pathway targets for potential therapeutic interventions to delay or avoid fibrosis initiation in TM and LC tissues.

Bioinformatics Meets Biomedicine: OncoFinder, a Quantitative Approach for Interrogating Molecular Pathways Using Gene Expression Data

We propose a biomathematical approach termed OncoFinder (OF) that enables performing both quantitative and qualitative analyses of the intracellular molecular pathway activation. OF utilizes an algorithm that distinguishes the activator/repressor role of every gene product in a pathway. This method is applicable for the analysis of any physiological, stress, malignancy, and other conditions at the molecular level. OF showed a strong potential to neutralize background-caused differences between experimental gene expression data obtained using NGS, microarray and modern proteomics techniques. Importantly, in most cases, pathway activation signatures were better markers of cancer progression compared to the individual gene products. OF also enables correlating pathway activation with the success of anticancer therapy for individual patients. We further expanded this approach to analyze impact of micro RNAs (miRs) on the regulation of cellular interactome. Many alternative sources provide information about miRs and their targets. However, instruments elucidating higher level impact of the established total miR profiles are still largely missing. A variant of OncoFinder termed MiRImpact enables linking miR expression data with its estimated outcome on the regulation of molecular processes, such as signaling, metabolic, cytoskeleton, and DNA repair pathways. MiRImpact was used to establish cancer-specific and cytomegaloviral infection-linked interactomic signatures for hundreds of molecular pathways. Interestingly, the impact of miRs appeared orthogonal to pathway regulation at the mRNA level, which stresses the importance of combining all available levels of gene regulation to build a more objective molecular model of cell.

Mathematical Justification of Expression-Based Pathway Activation Scoring (PAS)

Although modeling of activation kinetics for various cell signaling pathways has reached a high grade of sophistication and thoroughness, most such kinetic models still remain of rather limited practical value for biomedicine. Nevertheless, recent advancements have been made in application of signaling pathway science for real needs of prescription of the most effective drugs for individual patients. The methods for such prescription evaluate the degree of pathological changes in the signaling machinery based on two types of data: first, on the results of high-throughput gene expression profiling, and second, on the molecular pathway graphs that reflect interactions between the pathway members. For example, our algorithm OncoFinder evaluates the activation of molecular pathways on the basis of gene/protein expression data in the objects of the interest.Yet, the question of assessment of the relative importance for each gene product in a molecular pathway remains unclear unless one call for the methods of parameter sensitivity /stiffness analysis in the interactomic kinetic models of signaling pathway activation in terms of total concentrations of each gene product.Here we show two principal points: 1. First, the importance coefficients for each gene in pathways that were obtained using the extremely time- and labor-consuming stiffness analysis of full-scaled kinetic models generally differ from much easier-to-calculate expression-based pathway activation score (PAS) not more than by 30%, so the concept of PAS is kinetically justified. 2. Second, the use of pathway-based approach instead of distinct gene analysis, due to the law of large numbers, allows restoring the correlation between the similar samples that were examined using different transcriptome investigation techniques.

MiRImpact, a new bioinformatic method using complete microRNA expression profiles to assess their overall influence on the activity of intracellular molecular pathways

MicroRNAs (miRs) are short noncoding RNA molecules that regulate expression of target mRNAs. Many published sources provide information about miRs and their targets. However, bioinformatic tools elucidating higher level impact of the established total miR profiles, are still largely missing. Recently, we developed a method termed OncoFinder enabling quantification of the activities of intracellular molecular pathways basing on gene expression data. Here we propose a new technique, MiRImpact, which enables to link miR expression data with its estimated outcome on the regulation of molecular pathways, like signaling, metabolic, cytoskeleton rearrangement, and DNA repair pathways. MiRImpact uses OncoFinder rationale for pathway activity calculations, with the major distinctions that (i) it deals with the concentrations of miRs - known regulators of gene products participating in molecular pathways, and (ii) miRs are considered as negative regulators of target molecules, if other is not specified. MiRImpact operates with 2 types of databases: for molecular targets of miRs and for gene products participating in molecular pathways. We applied MiRImpact to compare regulation of human bladder cancer-specific signaling pathways at the levels of mRNA and miR expression. We took 2 most complete alternative databases of experimentally validated miR targets – miRTarBase and DianaTarBase, and an OncoFinder database featuring 2725 gene products and 271 signaling pathways. We showed that the impact of miRs is orthogonal to pathway regulation at the mRNA level, which stresses the importance of studying posttranscriptional regulation of gene expression. We also report characteristic set of miR and mRNA regulation features linked with bladder cancer.

Towards understanding the breast cancer epigenome: a comparison of genome-wide DNA methylation and gene expression data

Until recently, an elevated disease risk has been ascribed to a genetic predisposition, however, exciting progress over the past years has discovered alternate elements of inheritance that involve epigenetic regulation. Epigenetic changes are heritably stable alterations that include DNA methylation, histone modifications and RNA-mediated silencing. Aberrant DNA methylation is a common molecular basis for a number of important human diseases, including breast cancer. Changes in DNA methylation profoundly affect global gene expression patterns. What is emerging is a more dynamic and complex association between DNA methylation and gene expression than previously believed. Although many tools have already been developed for analyzing genome-wide gene expression data, tools for analyzing genome-wide DNA methylation have not yet reached the same level of refinement.

Here we provide an in-depth analysis of DNA methylation in parallel with gene expression data characteristics and describe the particularities of low-level and high-level analyses of DNA methylation data. Low-level analysis refers to pre-processing of methylation data (i.e. normalization, transformation and filtering), whereas high-level analysis is focused on illustrating the application of the widely used class comparison, class prediction and class discovery methods to DNA methylation data. Furthermore, we investigate the influence of DNA methylation on gene expression by measuring the correlation between the degree of CpG methylation and the level of expression and to explore the pattern of methylation as a function of the promoter region.

Molecular pathway activation features linked with transition from normal skin to primary and metastatic melanomas in human

Melanoma is the most aggressive and dangerous type of skin cancer, but its molecular mechanisms remain largely unclear. For transcriptomic data of 478 primary and metastatic melanoma, nevi and normal skin samples, we performed high-throughput analysis of intracellular molecular networks including 592 signaling and metabolic pathways. We showed that at the molecular pathway level, the formation of nevi largely resembles transition from normal skin to primary melanoma. Using a combination of bioinformatic machine learning algorithms, we identified 44 characteristic signaling and metabolic pathways connected with the formation of nevi, development of primary melanoma, and its metastases. We created a model describing formation and progression of melanoma at the level of molecular pathway activation. We discovered six novel associations between activation of metabolic molecular pathways and progression of melanoma: for allopregnanolone biosynthesis, L-carnitine biosynthesis, zymosterol biosynthesis (inhibited in melanoma), fructose 2, 6-bisphosphate synthesis and dephosphorylation, resolvin D biosynthesis (activated in melanoma), D-myo-inositol hexakisphosphate biosynthesis (activated in primary, inhibited in metastatic melanoma). Finally, we discovered fourteen tightly coordinated functional clusters of molecular pathways. This study helps to decode molecular mechanisms underlying the development of melanoma.

Combinatorial high-throughput experimental and bioinformatic approach identifies molecular pathways linked with the sensitivity to anticancer target drugs

Effective choice of anticancer drugs is important problem of modern medicine. We developed a method termed OncoFinder for the analysis of new type of biomarkers reflecting activation of intracellular signaling and metabolic molecular pathways. These biomarkers may be linked with the sensitivity to anticancer drugs. In this study, we compared the experimental data obtained in our laboratory and in the Genomics of Drug Sensitivity in Cancer (GDS) project for testing response to anticancer drugs and transcriptomes of various human cell lines. The microarray-based profiling of transcriptomes was performed for the cell lines before the addition of drugs to the medium, and experimental growth inhibition curves were built for each drug, featuring characteristic IC₅₀ values. We assayed here four target drugs - Pazopanib, Sorafenib, Sunitinib and Temsirolimus, and 238 different cell lines, of which 11 were profiled in our laboratory and 227 - in GDS project. Using the OncoFinder-processed transcriptomic data on ∼600 molecular pathways, we identified pathways showing significant correlation between pathway activation strength (PAS) and IC₅₀ values for these drugs. Correlations reflect relationships between response to drug and pathway activation features. We intersected the results and found molecular pathways significantly correlated in both our assay and GDS project. For most of these pathways, we generated molecular models of their interaction with known molecular target(s) of the respective drugs. For the first time, our study uncovered mechanisms underlying cancer cell response to drugs at the high-throughput molecular interactomic level.

A method for predicting target drug efficiency in cancer based on the analysis of signaling pathway activation

A new generation of anticancer therapeutics called target drugs has quickly developed in the 21st century. These drugs are tailored to inhibit cancer cell growth, proliferation, and viability by specific interactions with one or a few target proteins. However, despite formally known molecular targets for every "target" drug, patient response to treatment remains largely individual and unpredictable. Choosing the most effective personalized treatment remains a major challenge in oncology and is still largely trial and error. Here we present a novel approach for predicting target drug efficacy based on the gene expression signature of the individual tumor sample(s). The enclosed bioinformatic algorithm detects activation of intracellular regulatory pathways in the tumor in comparison to the corresponding normal tissues. According to the nature of the molecular targets of a drug, it predicts whether the drug can prevent cancer growth and survival in each individual case by blocking the abnormally activated tumor-promoting pathways or by reinforcing internal tumor suppressor cascades. To validate the method, we compared the distribution of predicted drug efficacy scores for five drugs (Sorafenib, Bevacizumab, Cetuximab, Sorafenib, Imatinib, Sunitinib) and seven cancer types (Clear Cell Renal Cell Carcinoma, Colon cancer, Lung adenocarcinoma, non-Hodgkin Lymphoma, Thyroid cancer and Sarcoma) with the available clinical trials data for the respective cancer types and drugs. The percent of responders to a drug treatment correlated significantly (Pearson's correlation 0.77 p = 0.023) with the percent of tumors showing high drug scores calculated with the current algorithm.

Quantifying signaling pathway activation to monitor the quality of induced pluripotent stem cells

Many attempts have been made to evaluate the safety and potency of human induced pluripotent stem cells (iPSCs) for clinical applications using transcriptome data, but results so far have been ambiguous or even contradictory. Here, we characterized stem cells at the pathway level, rather than at the gene level as has been the focus of previous work. We meta-analyzed publically-available gene expression data sets and evaluated signaling and metabolic pathway activation profiles for 20 human embryonic stem cell (ESC) lines, 12 human iPSC lines, five embryonic body lines, and six fibroblast cell lines. We demonstrated the close resemblance of iPSCs with ESCs at the pathway level, and provided examples of how pathway activity can be applied to identify iPSC line abnormalities or to predict in vitro differentiation potential. Our results indicate that pathway activation profiling is a promising strategy for evaluating the safety and potency of iPSC lines in translational medicine applications.

Common pathway signature in lung and liver fibrosis

Fibrosis, a progressive accumulation of extracellular matrix components, encompasses a wide spectrum of distinct organs, and accounts for an increasing burden of morbidity and mortality worldwide. Despite the tremendous clinical impact, the mechanisms governing the fibrotic process are not yet understood, and to date, no clinically reliable therapies for fibrosis have been discovered. Here we applied Regeneration Intelligence, a new bioinformatics software suite for qualitative analysis of intracellular signaling pathway activation using transcriptomic data, to assess a network of molecular signaling in lung and liver fibrosis. In both tissues, our analysis detected major conserved signaling pathways strongly associated with fibrosis, suggesting that some of the pathways identified by our algorithm but not yet wet-lab validated as fibrogenesis related, may be attractive targets for future research. While the majority of significantly disrupted pathways were specific to histologically distinct organs, several pathways have been concurrently activated or downregulated among the hepatic and pulmonary fibrosis samples, providing new evidence of evolutionary conserved pathways that may be relevant as possible therapeutic targets. While future confirmatory studies are warranted to validate these observations, our platform proposes a promising new approach for detecting fibrosis-promoting pathways and tailoring the right therapy to prevent fibrogenesis.

Applications of Deep Learning in Biomedicine

Increases in throughput and installed base of biomedical research equipment led to a massive accumulation of -omics data known to be highly variable, high-dimensional, and sourced from multiple often incompatible data platforms. While this data may be useful for biomarker identification and drug discovery, the bulk of it remains underutilized. Deep neural networks (DNNs) are efficient algorithms based on the use of compositional layers of neurons, with advantages well matched to the challenges -omics data presents. While achieving state-of-the-art results and even surpassing human accuracy in many challenging tasks, the adoption of deep learning in biomedicine has been comparatively slow. Here, we discuss key features of deep learning that may give this approach an edge over other machine learning methods. We then consider limitations and review a number of applications of deep learning in biomedical studies demonstrating proof of concept and practical utility.

ROMA: Representation and Quantification of Module Activity from Target Expression Data

In many analyses of high-throughput data in systems biology, there is a need to quantify the activity of a set of genes in individual samples. A typical example is the case where it is necessary to estimate the activity of a transcription factor (which is often not directly measurable) from the expression of its target genes. We present here ROMA (Representation and quantification Of Module Activities) Java software, designed for fast and robust computation of the activity of gene sets (or modules) with coordinated expression. ROMA activity quantification is based on the simplest uni-factor linear model of gene regulation that approximates the expression data of a gene set by its first principal component. The proposed algorithm implements novel functionalities: it provides several method modifications for principal components computation, including weighted, robust and centered methods; it distinguishes overdispersed modules (based on the variance explained by the first principal component) and coordinated modules (based on the significance of the spectral gap); finally, it computes statistical significance of the estimated module overdispersion or coordination. ROMA can be applied in many contexts, from estimating differential activities of transcriptional factors to finding overdispersed pathways in single-cell transcriptomics data. We describe here the principles of ROMA providing several practical examples of its use. ROMA source code is available at https://github.com/sysbio-curie/Roma.