Dataset Information

Integrative machine learning analysis of multiple gene expression profiles in cervical cancer.

ABSTRACT: Although most of the cervical cancer cases are reported to be closely related to the Human Papillomavirus (HPV) infection, there is a need to study genes that stand up differentially in the final actualization of cervical cancers following HPV infection. In this study, we proposed an integrative machine learning approach to analyse multiple gene expression profiles in cervical cancer in order to identify a set of genetic markers that are associated with and may eventually aid in the diagnosis or prognosis of cervical cancers. The proposed integrative analysis is composed of three steps: namely, (i) gene expression analysis of individual dataset; (ii) meta-analysis of multiple datasets; and (iii) feature selection and machine learning analysis. As a result, 21 gene expressions were identified through the integrative machine learning analysis which including seven supervised and one unsupervised methods. A functional analysis with GSEA (Gene Set Enrichment Analysis) was performed on the selected 21-gene expression set and showed significant enrichment in a nine-potential gene expression signature, namely PEG3, SPON1, BTD and RPLP2 (upregulated genes) and PRDX3, COPB2, LSM3, SLC5A3 and AS1B (downregulated genes).

SUBMITTER: Tan MS

PROVIDER: S-EPMC6064203 | biostudies-literature | 2018

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Integrative machine learning analysis of multiple gene expression profiles in cervical cancer.

Tan Mei Sze MS Chang Siow-Wee SW Cheah Phaik Leng PL Yap Hwa Jen HJ

PeerJ 20180725

Although most of the cervical cancer cases are reported to be closely related to the Human Papillomavirus (HPV) infection, there is a need to study genes that stand up differentially in the final actualization of cervical cancers following HPV infection. In this study, we proposed an integrative machine learning approach to analyse multiple gene expression profiles in cervical cancer in order to identify a set of genetic markers that are associated with and may eventually aid in the diagnosis or ...[more]

PMID: 30065881

Similar Datasets

Project description:Tumor progression includes the obtainment of progenitor and stem cell-like features and the gradual loss of a differentiated phenotype. Stemness was defined as the potential for differentiation and self-renewal from the cell of origin. Previous studies have confirmed the effective application of stemness in a number of malignancies. However, the mechanisms underlying the growth and maintenance of multiple myeloma (MM) stem cells remain unclear. We calculated the stemness index for samples of MM by utilizing a novel one-class logistic regression (OCLR) machine learning algorithm and found that mRNA expression-based stemness index (mRNAsi) was an independent prognostic factor of MM. Based on the same cutoff value, mRNAsi could stratify MM patients into low and high groups with different outcomes. We identified 127 stemness-related signatures using weighted gene co-expression network analysis (WGCNA) and differential expression analysis. Functional annotation and pathway enrichment analysis indicated that these genes were mainly involved in the cell cycle, cell differentiation, and DNA replication and repair. Using the molecular complex detection (MCODE) algorithm, we identified 34 pivotal signatures. Meanwhile, we conducted unsupervised clustering and classified the MM cohorts into three MM stemness (MMS) clusters with distinct prognoses. Samples in MMS-cluster3 possessed the highest stemness fractions and the worst prognosis. Additionally, we applied the ESTIMATE algorithm to infer differential immune infiltration among the three MMS clusters. The immune core and stromal score were significantly lower in MMS-cluster3 than in the other clusters, supporting the negative relation between stemness and anticancer immunity. Finally, we proposed a prognostic nomogram that allows for individualized assessment of the 3- and 5-year overall survival (OS) probabilities among patients with MM. Our study comprehensively assessed the MM stemness index based on large cohorts and built a 34-gene based classifier for predicting prognosis and potential strategies for stemness treatment.

Project description:Recent advances in high-throughput technology and the emergence of large-scale genomic datasets have enabled detection of genomic features that affect clinical outcomes. Although many previous computational studies have analysed the effect of each single gene or the additive effects of multiple genes on the clinical outcome, less attention has been devoted to the identification of gene-gene interactions of general type that are associated with the clinical outcome. Moreover, the integration of information from multiple molecular profiles adds another challenge to this problem. Recently, network-based approaches have gained huge popularity. However, previous network construction methods have been more concerned with the relationship between features only, rather than the effect of feature interactions on clinical outcome.We propose a mutual information-based integrative network analysis framework (MINA) that identifies gene pairs associated with clinical outcome and systematically analyses the resulting networks over multiple genomic profiles. We implement an efficient non-parametric testing scheme that ensures the significance of detected gene interactions. We develop a tool named MINA that automates the proposed analysis scheme of identifying outcome-associated gene interactions and generating various networks from those interacting pairs for downstream analysis.We demonstrate the proposed framework using real data from ovarian cancer patients in The Cancer Genome Atlas (TCGA). Statistically significant gene pairs associated with survival were identified from multiple genomic profiles, which include many individual genes that have weak or no effect on survival. Moreover, we also show that integrated networks, constructed by merging networks from multiple genomic profiles, demonstrate better topological properties and biological significance than individual networks.We have developed a simple but powerful analysis tool that is able to detect gene-gene interactions associated with clinical outcome on multiple genomic profiles. By being network-based, our approach provides a better insight into the underlying gene-gene interaction mechanisms that affect the clinical outcome of cancer patients.

Project description:Cervical carcinoma is a global public health burden. Given that it is usually asymptomatic at potentially curative stages, the development of clinically accurate tests is critical for early detection and individual risk stratification. The present study performed an integrative meta-analysis of the transcriptomes from 10 cervical carcinoma cohorts, with the aim of identifying biomarkers that are associated with malignant transformation of cervical epithelium, and establish their clinical applicability. From among the top ranked differentially expressed genes, flap structure-specific endonuclease 1 (FEN1) and poly (U)-specific endoribonuclease (ENDOU) were selected for further validation, and their clinical applicability was assessed using immunohistochemically stained microarrays comprising 110 tissue cores, using p16 and Ki67 staining as the comparator tests. The results demonstrated that FEN1 expression was significantly upregulated in 65% of tumor specimens (P=0.0001), with no detectable expression in the non-tumor tissues. Furthermore, its expression was significantly associated with Ki67 staining in tumor samples (P<0.0001), but no association was observed with p16 expression or the presence of human papilloma virus types 16/18, patient age, tumor grade or stage. FEN1 staining demonstrated lower sensitivity than p16 (69.3 vs. 96.8%) and Ki67 (69.3 vs. 76.3%); however, the specificity was identical to p16 and higher than that of Ki67 (100 vs. 71.4%).ENDOU staining was consistent with the microarray results, demonstrating 1% positivity in tumors and 40% positivity in non-tumor tissues. Gene set enrichment analysis of cervical tumors overexpressing FEN1 revealed its association with enhanced growth factor signaling, immune response inhibition and extracellular matrix remodeling, whereas tumors with low ENDOU expression exhibited inhibition of epithelial development and differentiation processes. Taken together, the results of the present study demonstrate the feasibility of the integrative meta-analysis approach to identify relevant biomarkers associated with cervical carcinogenesis. Thus, FEN1 and ENDOU may be useful diagnostic biomarkers for squamous cervical carcinoma. However, further studies are required to determine their diagnostic performance in larger patient cohorts and validate the results presented here.

Dataset Information

Integrative machine learning analysis of multiple gene expression profiles in cervical cancer.

Publications

Integrative machine learning analysis of multiple gene expression profiles in cervical cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets