Dataset Information

G2Vec: Distributed gene representations for identification of cancer prognostic genes.

ABSTRACT: Identification of cancer prognostic genes is important in that it can lead to accurate outcome prediction and better therapeutic trials for cancer patients. Many computational approaches have been proposed to achieve this goal; however, there is room for improvement. Recent developments in deep learning techniques can aid in the identification of better prognostic genes and more accurate outcome prediction, but one of the main problems in the adoption of deep learning for this purpose is that data from cancer patients have too many dimensions, while the number of samples is relatively small. In this study, we propose a novel network-based deep learning method to identify prognostic gene signatures via distributed gene representations generated by G2Vec, which is a modified Word2Vec model originally used for natural language processing. We applied the proposed method to five cancer types including liver cancer and showed that G2Vec outperformed extant feature selection methods, especially for small number of samples. Moreover, biomarkers identified by G2Vec was useful to find significant prognostic gene modules associated with hepatocellular carcinoma.

SUBMITTER: Choi J

PROVIDER: S-EPMC6137174 | biostudies-literature | 2018 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

G2Vec: Distributed gene representations for identification of cancer prognostic genes.

Choi Jonghwan J Oh Ilhwan I Seo Sangmin S Ahn Jaegyoon J

Scientific reports 20180913 1

Identification of cancer prognostic genes is important in that it can lead to accurate outcome prediction and better therapeutic trials for cancer patients. Many computational approaches have been proposed to achieve this goal; however, there is room for improvement. Recent developments in deep learning techniques can aid in the identification of better prognostic genes and more accurate outcome prediction, but one of the main problems in the adoption of deep learning for this purpose is that da ...[more]

PMID: 30213980

Similar Datasets

Project description:Background and Objective: Bladder cancer is the most common tumor in the urinary system, with a higher incidence in men than in women and a high recurrence rate. However, the mechanism of recurrence is still unclear, and it is urgent to clarify the pathophysiological mechanism of bladder cancer. To provide theoretical basis for the development of new therapies, investigating the effect of tumor microenvironment on the prognosis of bladder cancer is necessary. Methods: We applied the Estimation of STromal and Immune cells in MAlignant Tumors using Expression data (ESTIMATE) algorithm to the downloaded TCGA (The Cancer Genome Atlas) transcriptome data to obtain the immune scores and stromal scores of each sample, and then divided the samples into two groups: high and low immune scores (or high and low stromal scores), and found that some differential genes were associated with poor prognosis of patients. We then performed protein-protein interaction (PPI) network analysis to explore the relationship between these differentially expressed genes. Moreover, we also performed (Gene Ontology) GO and (Kyoto Encyclopedia of Genes and Genomes) KEGG analyses to explore the potential functions of differentially expressed genes. Finally, our results were validated in an independent dataset. Results: We identified 136 tumor microenvironment-related genes associated with poor prognosis of bladder cancer. GO annotation and KEGG pathway enrichment analysis found that these genes are mainly involved in extracellular matrix, Focal adhesion and phosphatidylinositol 3 kinase-protein kinaseB (PI3k-Akt) signaling pathway. Next, PPI network analysis revealed some hub genes including Versican (VCAN), Thrombospondin 1 (THBS1) and Thrombospondin 1 (THBS2). Finally, 27 genes were further verified in the independent data set. Conclusions: We found 27 tumor microenvironment-related genes of bladder cancer, which are associated with poor prognosis of bladder cancer. These genes may inspire researchers to develop new treatments for bladder cancer.

Dataset Information

G2Vec: Distributed gene representations for identification of cancer prognostic genes.

Publications

G2Vec: Distributed gene representations for identification of cancer prognostic genes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets