Network-based stratification of tumor mutations.
Ontology highlight
ABSTRACT:
SUBMITTER: Hofree M
PROVIDER: S-EPMC3866081 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:MotivationTumor stratification has a wide range of biomedical and clinical applications, including diagnosis, prognosis and personalized treatment. However, cancer is always driven by the combination of mutated genes, which are highly heterogeneous across patients. Accurately subdividing the tumors into subtypes is challenging.ResultsWe developed a network-embedding based stratification (NES) methodology to identify clinically relevant patient subtypes from large-scale patients' somatic mutation profiles. The central hypothesis of NES is that two tumors would be classified into the same subtypes if their somatic mutated genes located in the similar network regions of the human interactome. We encoded the genes on the human protein-protein interactome with a network embedding approach and constructed the patients' vectors by integrating the somatic mutation profiles of 7344 tumor exomes across 15 cancer types. We firstly adopted the lightGBM classification algorithm to train the patients' vectors. The AUC value is around 0.89 in the prediction of the patient's cancer type and around 0.78 in the prediction of the tumor stage within a specific cancer type. The high classification accuracy suggests that network embedding-based patients' features are reliable for dividing the patients. We conclude that we can cluster patients with a specific cancer type into several subtypes by using an unsupervised clustering algorithm to learn the patients' vectors. Among the 15 cancer types, the new patient clusters (subtypes) identified by the NES are significantly correlated with patient survival across 12 cancer types. In summary, this study offers a powerful network-based deep learning methodology for personalized cancer medicine.Availability and implementationSource code and data can be downloaded from https://github.com/ChengF-Lab/NES.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC8034530 | biostudies-literature
Project description:Hepatocellular carcinoma (HCC) is one of the most frequent forms of liver cancer, and effective treatment methods are limited due to tumor heterogeneity. There is a great need for comprehensive approaches to stratify HCC patients, gain biological insights into subtypes, and ultimately identify effective therapeutic targets. We stratified HCC patients and characterized each subtype using transcriptomics data, genome-scale metabolic networks and network topology/controllability analysis. This comprehensive systems-level analysis identified three distinct subtypes with substantial differences in metabolic and signaling pathways reflecting at genomic, transcriptomic, and proteomic levels. These subtypes showed large differences in clinical survival associated with altered kynurenine metabolism, WNT/β-catenin-associated lipid metabolism, and PI3K/AKT/mTOR signaling. Integrative analyses indicated that the three subtypes rely on alternative enzymes (e.g., ACSS1/ACSS2/ACSS3, PKM/PKLR, ALDOB/ALDOA, MTHFD1L/MTHFD2/MTHFD1) to catalyze the same reactions. Based on systems-level analysis, we identified 8 to 28 subtype-specific genes with pivotal roles in controlling the metabolic network and predicted that these genes may be targeted for development of treatment strategies for HCC subtypes by performing in silico analysis. To validate our predictions, we performed experiments using HepG2 cells under normoxic and hypoxic conditions and observed opposite expression patterns between genes expressed in high/moderate/low-survival tumor groups in response to hypoxia, reflecting activated hypoxic behavior in patients with poor survival. In conclusion, our analyses showed that the heterogeneous HCC tumors can be stratified using a metabolic network-driven approach, which may also be applied to other cancer types, and this stratification may have clinical implications to drive the development of precision medicine.
| S-EPMC6294939 | biostudies-literature
Project description:(1) Background: The identification of tumor subtypes is fundamental in precision medicine for accurate diagnoses and personalized therapies. Cancer development is often driven by the accumulation of somatic mutations that can cause alterations in tissue functions and morphologies. In this work, a method based on a deep neural network integrated into a network-based stratification framework (D3NS) is proposed to stratify tumors according to somatic mutations. (2) Methods: This approach leverages the power of deep neural networks to detect hidden information in the data by combining the knowledge contained in a network of gene interactions, as typical of network-based stratification methods. D3NS was applied using real-world data from The Cancer Genome Atlas for bladder, ovarian, and kidney cancers. (3) Results: This technique allows for the identification of tumor subtypes characterized by different survival rates and significant associations with several clinical outcomes (tumor stage, grade or response to therapy). (4) Conclusion: D3NS can provide a base model in cancer research and could be considered as a useful tool for tumor stratification, offering potential support in clinical settings.
| S-EPMC11352240 | biostudies-literature
Project description:BackgroundCancers are complex diseases with heterogeneous genetic causes and clinical outcomes. It is critical to classify patients into subtypes and associate the subtypes with clinical outcomes for better prognosis and treatment. Large-scale studies have comprehensively identified somatic mutations across multiple tumor types, providing rich datasets for classifying patients based on genomic mutations. One challenge associated with this task is that mutations are rarely shared across patients. Network-based stratification (NBS) approaches have been proposed to overcome this challenge and used to classify tumors based on exome-level mutations. In routine research and clinical applications, however, usually only a small panel of pre-selected genes is screened for mutations. It is unknown whether such small panels are effective in classifying patients into clinically meaningful subtypes.ResultsIn this study, we applied NBS to 13 major cancer types with exome-level mutation data and compared the classification based on the full exome data with those focusing only on small sets of genes. Specifically, we investigated three panels, FoundationOne (240 genes), PanCan (127 genes) and TruSeq (48 genes). We showed that small panels not only are effective in clustering tumors but also often outperform full exome data for most cancer types. We further associated subtypes with clinical data and identified 5 tumor types (CRC-Colorectal carcinoma, HNSC-Head and neck squamous cell carcinoma, KIRC-Kidney renal clear cell carcinoma, LUAD-Lung adenocarcinoma and UCEC-Uterine corpus endometrial carcinoma) whose subtypes are significantly associated with overall survival, all based on small panels.ConclusionOur analyses indicate that effective patient subtyping can be carried out using mutations detected in smaller gene panels, probably due to the enrichment of clinically important genes in such panels.
| S-EPMC4474538 | biostudies-literature
Project description:Cancer is a complex disease involving multiple genomic alterations that disrupt the dynamic response of signaling networks. The heterogeneous nature of cancer, which results in highly variable drug response, is a major obstacle to developing effective cancer therapy. Previous studies of cancer therapeutic response mostly focus on static analysis of genome-wide alterations, thus they are unable to unravel the dynamic, network-specific origin of variation. Here we present a network dynamics-based approach to integrate cancer genomics with dynamics of biological network for drug response prediction and design of drug combination. We select the p53 network as an example and analyze its cancer-specific state transition dynamics under distinct anticancer drug treatments by attractor landscape analysis. Our results not only enable stratification of cancer into distinct drug response groups, but also reveal network-specific drug targets that maximize p53 network-mediated cell death, providing a basis to design combinatorial therapeutic strategies for distinct cancer genomic subtypes.
| S-EPMC5717260 | biostudies-literature
Project description:The stratification of cancer into subtypes that are significantly associated with clinical outcomes is beneficial for targeted prognosis and treatment. In this study, we integrated somatic mutation and gene expression data to identify clusters of patients. In contrast to previous studies, we constructed cancer-type-specific significant co-expression networks (SCNs) rather than using a fixed gene network across all cancers, such as the network-based stratification (NBS) method, which ignores cancer heterogeneity. For each type of cancer, the gene expression data were used to construct the SCN network, while the gene somatic mutation data were mapped onto the network, propagated, and used for further clustering. For the clustering, we adopted an improved network-regularized non-negative matrix factorization (netNMF) (netNMF_HC) for a more precise classification. We applied our method to various datasets, including ovarian cancer (OV), lung adenocarcinoma (LUAD) and uterine corpus endometrial carcinoma (UCEC) cohorts derived from the TCGA (The Cancer Genome Atlas) project. Based on the results, we evaluated the performance of our method to identify survival-relevant subtypes and further compared it to the NBS method, which adopts priori networks and netNMF algorithm. The proposed algorithm outperformed the NBS method in identifying informative cancer subtypes that were significantly associated with clinical outcomes in most cancer types we studied. In particular, our method identified survival-associated UCEC subtypes that were not identified by the NBS method. Our analysis indicated valid subtyping of patient could be applied by mutation data with cancer-type-specific SCNs and netNMF_HC for individual cancers because of specific cancer co-expression patterns and more precise clustering.
| S-EPMC5433734 | biostudies-literature
Project description:BackgroundA major challenge of bioinformatics in the era of precision medicine is to identify the molecular biomarkers for complex diseases. It is a general expectation that these biomarkers or signatures have not only strong discrimination ability, but also readable interpretations in a biological sense. Generally, the conventional expression-based or network-based methods mainly capture differential genes or differential networks as biomarkers, however, such biomarkers only focus on phenotypic discrimination and usually have less biological or functional interpretation. Meanwhile, the conventional function-based methods could consider the biomarkers corresponding to certain biological functions or pathways, but ignore the differential information of genes, i.e., disregard the active degree of particular genes involved in particular functions, thereby resulting in less discriminative ability on phenotypes. Hence, it is strongly demanded to develop elaborate computational methods to directly identify functional network biomarkers with both discriminative power on disease states and readable interpretation on biological functions.ResultsIn this paper, we present a new computational framework based on an integer programming model, named as Comparative Network Stratification (CNS), to extract functional or interpretable network biomarkers, which are of strongly discriminative power on disease states and also readable interpretation on biological functions. In addition, CNS can not only recognize the pathogen biological functions disregarded by traditional Expression-based/Network-based methods, but also uncover the active network-structures underlying such dysregulated functions underestimated by traditional Function-based methods. To validate the effectiveness, we have compared CNS with five state-of-the-art methods, i.e. GSVA, Pathifier, stSVM, frSVM and AEP on four datasets of different complex diseases. The results show that CNS can enhance the discriminative power of network biomarkers, and further provide biologically interpretable information or disease pathogenic mechanism of these biomarkers. A case study on type 1 diabetes (T1D) demonstrates that CNS can identify many dysfunctional genes and networks previously disregarded by conventional approaches.ConclusionTherefore, CNS is actually a powerful bioinformatics tool, which can identify functional or interpretable network biomarkers with both discriminative power on disease states and readable interpretation on biological functions. CNS was implemented as a Matlab package, which is available at http://www.sysbio.ac.cn/cb/chenlab/images/CNSpackage_0.1.rar .
| S-EPMC5374559 | biostudies-literature
Project description:The tumor immune microenvironment (TIME) plays an important role in penile squamous cell carcinoma (peSCC) pathogenesis. Here, the immunophenotype of the TIME in peSCC was determined by integrating the expression patterns of immune checkpoints (programmed cell death-1 (PD-1)/programmed cell death ligand-1 (PD-L1), cytotoxic T lymphocyte antigen 4 (CTLA-4), and Siglec-15) and the components of tumor-infiltrating lymphocytes, including CD8+ or Granzyme B+ T cells, FOXP3+ regulatory T cells, and CD68+ or CD206+ macrophages, in 178 patients. A high density of Granzyme B, FOXP3, CD68, CD206, PD-1, and CTLA-4 was associated with better disease-specific survival (DSS). The patients with diffuse PD-L1 tumor cell expression had worse prognoses than those with marginal or negative PD-L1 expression. Four immunophenotypes were identified by unsupervised clustering analysis, based on certain immune markers, which were associated with DSS and lymph node metastasis (LNM) in peSCC. There was no significant relationship between the immunophenotypes and high-risk human papillomavirus (hrHPV) infection. However, the hrHPV-positive peSCC exhibited a higher density of stromal Granzyme B and intratumoral PD-1 than the hrHPV-negative tumors (p = 0.049 and 0.002, respectively). In conclusion, the immunophenotypes of peSCC were of great value in predicting LNM and prognosis, and may provide support for clinical stratification management and immunotherapy intervention.
| S-EPMC7407624 | biostudies-literature
Project description:Melanoma is one of the most aggressive types of skin cancer, with significant heterogeneity in overall survival. Currently, tumor-node-metastasis (TNM) staging is insufficient to provide accurate survival prediction and appropriate treatment decision making for several types of tumors, such as those in melanoma patients. Therefore, the identification of more reliable prognosis biomarkers is urgently essential. Recent studies have shown that low immune cells infiltration is significantly associated with unfavorable clinical outcome in melanoma patients. Here we constructed a prognostic-related gene signature for melanoma risk stratification by quantifying the levels of several cancer hallmarks and identify the Wnt/β-catenin activation pathway as a primary risk factor for low tumor immunity. A series of bioinformatics and statistical methods were combined and applied to construct a Wnt-immune-related prognosis gene signature. With this gene signature, we computed risk scores for individual patients that can predict overall survival. To evaluate the robustness of the result, we validated the signature in multiple independent GEO datasets. Finally, an overall survival-related nomogram was established based on the gene signature and clinicopathological features. The Wnt-immune-related prognostic risk score could better predict overall survival compared with standard clinicopathological features. Our results provide a comprehensive map of the oncogene-immune-related gene signature that can serve as valuable biomarkers for better clinical decision making.
| S-EPMC8584987 | biostudies-literature
Project description:Risk stratification model for lung cancer with gene expression profile is of great interest. Instead of previous models based on individual prognostic genes, we aimed to develop a novel system-level risk stratification model for lung adenocarcinoma based on gene coexpression network. Using multiple microarray, gene coexpression network analysis was performed to identify survival-related networks. A deep learning based risk stratification model was constructed with representative genes of these networks. The model was validated in two test sets. Survival analysis was performed using the output of the model to evaluate whether it could predict patients' survival independent of clinicopathological variables. Five networks were significantly associated with patients' survival. Considering prognostic significance and representativeness, genes of the two survival-related networks were selected for input of the model. The output of the model was significantly associated with patients' survival in two test sets and training set (p < 0.00001, p < 0.0001 and p = 0.02 for training and test sets 1 and 2, resp.). In multivariate analyses, the model was associated with patients' prognosis independent of other clinicopathological features. Our study presents a new perspective on incorporating gene coexpression networks into the gene expression signature and clinical application of deep learning in genomic data science for prognosis prediction.
| S-EPMC5822793 | biostudies-literature