scholarly journals Identification of miRNA-Mediated Subpathways as Prostate Cancer Biomarkers Based on Topological Inference in a Machine Learning Process Using Integrated Gene and miRNA Expression Data

2021 ◽  
Vol 12 ◽  
Author(s):  
Ziyu Ning ◽  
Shuang Yu ◽  
Yanqiao Zhao ◽  
Xiaoming Sun ◽  
Haibin Wu ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Machine Learning ◽  
Single Molecule ◽  
Target Genes ◽  
Gene Expression Omnibus ◽  
The Cancer Genome Atlas ◽  
Differentially Expressed ◽  
Support Vector ◽  
Normal Prostate ◽  
Pathway Network

Accurately identifying classification biomarkers for distinguishing between normal and cancer samples is challenging. Additionally, the reproducibility of single-molecule biomarkers is limited by the existence of heterogeneous patient subgroups and differences in the sequencing techniques used to collect patient data. In this study, we developed a method to identify robust biomarkers (i.e., miRNA-mediated subpathways) associated with prostate cancer based on normal prostate samples and cancer samples from a dataset from The Cancer Genome Atlas (TCGA; n = 546) and datasets from the Gene Expression Omnibus (GEO) database (n = 139 and n = 90, with the latter being a cell line dataset). We also obtained 10 other cancer datasets to evaluate the performance of the method. We propose a multi-omics data integration strategy for identifying classification biomarkers using a machine learning method that involves reassigning topological weights to the genes using a directed random walk (DRW)-based method. A global directed pathway network (GDPN) was constructed based on the significantly differentially expressed target genes of the significantly differentially expressed miRNAs, which allowed us to identify the robust biomarkers in the form of miRNA-mediated subpathways (miRNAs). The activity value of each miRNA-mediated subpathway was calculated by integrating multiple types of data, which included the expression of the miRNA and the miRNAs’ target genes and GDPN topological information. Finally, we identified the high-frequency miRNA-mediated subpathways involved in prostate cancer using a support vector machine (SVM) model. The results demonstrated that we obtained robust biomarkers of prostate cancer, which could classify prostate cancer and normal samples. Our method outperformed seven other methods, and many of the identified biomarkers were associated with known clinical treatments.

Using machine learning method to identify MYLK as a novel marker to predict biochemical recurrence in prostate cancer

2021 ◽  
Vol 15 (1) ◽  
pp. 29-41
Author(s):  
Peng Qiao ◽  
Di Zhang ◽  
Song Zeng ◽  
Yicun Wang ◽  
Biao Wang ◽  
...  
Keyword(s):  
Gene Expression ◽  
Prostate Cancer ◽  
Biochemical Recurrence ◽  
Surgical Margin ◽  
Gene Expression Omnibus ◽  
Cancer Genome ◽  
The Cancer Genome Atlas ◽  
Support Vector ◽  
Cancer Genome Atlas ◽  
Genome Atlas

Aim: This study aims to identify novel marker to predict biochemical recurrence (BCR) in prostate cancer patients after radical prostatectomy with negative surgical margin. Materials & methods: The Cancer Genome Atlas database, Gene Expression Omnibus database and Cancer Cell Line Encyclopedia database were employed. The ensemble support vector machine-recursive feature elimination method was performed to select crucial gene for BCR. Results: We identified MYLK as a novel and independent biomarker for BCR in The Cancer Genome Atlas training cohort and confirmed in four independent Gene Expression Omnibus validation cohorts. Multi-omic analysis suggested that MYLK was a DNA methylation-driven gene. Additionally, MYLK had significant positive correlations with immune infiltrations. Conclusion: MYLK was identified and validated as a novel, robust and independent biomarker for BCR in prostate cancer.

scholarly journals MRI index lesion radiomics and machine learning for detection of extraprostatic extension of disease: a multicenter study

2021 ◽  
Author(s):  
Renato Cuocolo ◽  
Arnaldo Stanzione ◽  
Riccardo Faletti ◽  
Marco Gatti ◽  
Giorgio Calleris ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Machine Learning ◽  
Support Vector Machine ◽  
Support Vector ◽  
Index Lesion ◽  
Pairwise Correlation ◽  
Experienced Radiologist ◽  
Expert Radiologist ◽  
External Test ◽  
Extraprostatic Extension

Abstract Objectives To build a machine learning (ML) model to detect extraprostatic extension (EPE) of prostate cancer (PCa), based on radiomics features extracted from prostate MRI index lesions. Methods Consecutive MRI exams of patients undergoing radical prostatectomy for PCa were retrospectively collected from three institutions. Axial T2-weighted and apparent diffusion coefficient map images were annotated to obtain index lesion volumes of interest for radiomics feature extraction. Data from one institution was used for training, feature selection (using reproducibility, variance and pairwise correlation analyses, and a correlation-based subset evaluator), and tuning of a support vector machine (SVM) algorithm, with stratified 10-fold cross-validation. The model was tested on the two remaining institutions’ data and compared with a baseline reference and expert radiologist assessment of EPE. Results In total, 193 patients were included. From an initial dataset of 2436 features, 2287 were excluded due to either poor stability, low variance, or high collinearity. Among the remaining, 14 features were used to train the ML model, which reached an overall accuracy of 83% in the training set. In the two external test sets, the SVM achieved an accuracy of 79% and 74% respectively, not statistically different from that of the radiologist (81–83%, p = 0.39–1) and outperforming the baseline reference (p = 0.001–0.02). Conclusions A ML model solely based on radiomics features demonstrated high accuracy for EPE detection and good generalizability in a multicenter setting. Paired to qualitative EPE assessment, this approach could aid radiologists in this challenging task. Key Points • Predicting the presence of EPE in prostate cancer patients is a challenging task for radiologists. • A support vector machine algorithm achieved high diagnostic accuracy for EPE detection, with good generalizability when tested on multiple external datasets. • The performance of the algorithm was not significantly different from that of an experienced radiologist.

scholarly journals Characterization of Differentially Expressed miRNAs by CXCL12/SDF-1 in Human Bone Marrow Stromal Cells

2021 ◽  
Vol 12 (1) ◽  
pp. 132-143
Author(s):  
Matthew L. Potter ◽  
Kathryn Smith ◽  
Sagar Vyavahare ◽  
Sandeep Kumar ◽  
Sudharsan Periyasamy-Thandavan ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Fracture Healing ◽  
Osteogenic Differentiation ◽  
Stromal Cell ◽  
Target Genes ◽  
Fatty Acid Biosynthesis ◽  
Gene Expression Omnibus ◽  
Differentially Expressed ◽  
Mirna Regulation ◽  
Stromal Cell Derived Factor

Abstract Stromal cell-derived factor 1 (SDF-1) is known to influence bone marrow stromal cell (BMSC) migration, osteogenic differentiation, and fracture healing. We hypothesize that SDF-1 mediates some of its effects on BMSCs through epigenetic regulation, specifically via microRNAs (miRNAs). MiRNAs are small non-coding RNAs that target specific mRNA and prevent their translation. We performed global miRNA analysis and determined several miRNAs were differentially expressed in response to SDF-1 treatment. Gene Expression Omnibus (GEO) dataset analysis showed that these miRNAs play an important role in osteogenic differentiation and fracture healing. KEGG and GO analysis indicated that SDF-1 dependent miRNAs changes affect multiple cellular pathways, including fatty acid biosynthesis, thyroid hormone signaling, and mucin-type O-glycan biosynthesis pathways. Furthermore, bioinformatics analysis showed several miRNAs target genes related to stem cell migration and differentiation. This study's findings indicated that SDF-1 induces some of its effects on BMSCs function through miRNA regulation.

scholarly journals Identification of Key miRNAs in the Treatment of Dabrafenib-Resistant Melanoma

2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Guangyu Gao ◽  
Zhen Yao ◽  
Jiaofeng Shen ◽  
Yulong Liu
Keyword(s):  
Drug Resistance ◽  
Molecular Mechanism ◽  
Target Genes ◽  
Enrichment Analysis ◽  
Gene Expression Omnibus ◽  
The Cancer Genome Atlas ◽  
Pathway Enrichment Analysis ◽  
Ppi Network ◽  
Protein Protein Interaction ◽  
Obvious Association

Dabrafenib resistance is a significant problem in melanoma, and its underlying molecular mechanism is still unclear. The purpose of this study is to research the molecular mechanism of drug resistance of dabrafenib and to explore the key genes and pathways that mediate drug resistance in melanoma. GSE117666 was downloaded from the Gene Expression Omnibus (GEO) database and 492 melanoma statistics were also downloaded from The Cancer Genome Atlas (TCGA) database. Besides, differentially expressed miRNAs (DEMs) were identified by taking advantage of the R software and GEO2R. The Database for Annotation, Visualization, and Integrated Discovery (DAVID) and FunRich was used to perform Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis to identify potential pathways and functional annotations linked with melanoma chemoresistance. 9 DEMs and 872 mRNAs were selected after filtering. Then, target genes were uploaded to Metascape to construct protein-protein interaction (PPI) network. Also, 6 hub mRNAs were screened after performing the PPI network. Furthermore, a total of 4 out of 9 miRNAs had an obvious association with the survival rate ( P < 0.05 ) and showed a good power of risk prediction model of over survival. The present research may provide a deeper understanding of regulatory genes of dabrafenib resistance in melanoma.

scholarly journals Predicting lung adenocarcinoma disease progression using methylation-correlated blocks and ensemble machine learning classifiers

PeerJ ◽  
2021 ◽  
Vol 9 ◽  
pp. e10884
Author(s):  
Xin Yu ◽  
Qian Yang ◽  
Dong Wang ◽  
Zhaoyang Li ◽  
Nianhang Chen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Lung Adenocarcinoma ◽  
Cox Regression ◽  
Characteristic Curve ◽  
The Cancer Genome Atlas ◽  
Support Vector ◽  
Survival Prediction ◽  
Ensemble Model ◽  
Training Set ◽  
Cpg Sites

Applying the knowledge that methyltransferases and demethylases can modify adjacent cytosine-phosphorothioate-guanine (CpG) sites in the same DNA strand, we found that combining multiple CpGs into a single block may improve cancer diagnosis. However, survival prediction remains a challenge. In this study, we developed a pipeline named “stacked ensemble of machine learning models for methylation-correlated blocks” (EnMCB) that combined Cox regression, support vector regression (SVR), and elastic-net models to construct signatures based on DNA methylation-correlated blocks for lung adenocarcinoma (LUAD) survival prediction. We used methylation profiles from the Cancer Genome Atlas (TCGA) as the training set, and profiles from the Gene Expression Omnibus (GEO) as validation and testing sets. First, we partitioned the genome into blocks of tightly co-methylated CpG sites, which we termed methylation-correlated blocks (MCBs). After partitioning and feature selection, we observed different diagnostic capacities for predicting patient survival across the models. We combined the multiple models into a single stacking ensemble model. The stacking ensemble model based on the top-ranked block had the area under the receiver operating characteristic curve of 0.622 in the TCGA training set, 0.773 in the validation set, and 0.698 in the testing set. When stratified by clinicopathological risk factors, the risk score predicted by the top-ranked MCB was an independent prognostic factor. Our results showed that our pipeline was a reliable tool that may facilitate MCB selection and survival prediction.

scholarly journals Identification of microRNAs and genes as biomarkers of atrial fibrillation using a bioinformatics approach

2019 ◽  
Vol 47 (8) ◽  
pp. 3580-3589 ◽  
Author(s):  
Yingyuan Li ◽  
Wulin Tan ◽  
Fang Ye ◽  
Faling Xue ◽  
Shaowei Gao ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Target Genes ◽  
Enrichment Analysis ◽  
Gene Expression Omnibus ◽  
Differentially Expressed ◽  
Control Groups ◽  
Protein Protein Interaction ◽  
Differentially Expressed Mirnas ◽  
Software Modules ◽  
And Control

Objective We aimed to explore potential microRNAs (miRNAs) and target genes related to atrial fibrillation (AF). Methods Data for microarrays GSE70887 and GSE68475, both of which include AF and control groups, were downloaded from the Gene Expression Omnibus database. Differentially expressed miRNAs between AF and control groups were identified within each microarray, and the intersection of these two sets was obtained. These miRNAs were mapped to target genes in the miRNet database. Functional annotation and enrichment analysis of these target genes was performed in the DAVID database. The protein-protein interaction (PPI) network from the STRING database and the miRNA-target-gene network were merged into a PPI-miRNA network using Cytoscape software. Modules of this network containing miRNAs were detected and further analyzed. Results Ten differentially expressed miRNAs and 1520 target genes were identified. Three PPI-miRNA modules were constructed, which contained miR-424, miR-15a, miR-542-3p, and miR-421 as well as their target genes, CDK1, CDK6, and CCND3. Conclusion The identified miRNAs and genes may be related to the pathogenesis of AF. Thus, they may be potential biomarkers for diagnosis and targets for treatment of AF.

scholarly journals Analysis of PMEPA1 Isoforms (a and b) as Selective Inhibitors of Androgen and TGF-β Signaling Reveals Distinct Biological and Prognostic Features in Prostate Cancer

Cancers ◽  
2019 ◽  
Vol 11 (12) ◽  
pp. 1995
Author(s):  
Shashwat Sharad ◽  
Zsófia M. Sztupinszki ◽  
Yongmei Chen ◽  
Claire Kuo ◽  
Lakshmi Ravindranath ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Cell Proliferation ◽  
Transmembrane Protein ◽  
The Cancer Genome Atlas ◽  
Cell Counting ◽  
Luciferase Reporter ◽  
Mrna Levels ◽  
Normal Prostate ◽  
Prostate Tumorigenesis ◽  
Prognostic Features

Dysfunctions of androgen/TGF-β signaling play important roles in prostate tumorigenesis. Prostate Transmembrane Protein Androgen Induced 1 (PMEPA1) inhibits androgen and TGF-β signaling via a negative feedback loop. The loss of PMEPA1 confers resistance to androgen signaling inhibitors and promotes bone metastasis. Conflicting reports on the expression and biological functions of PMEPA1 in prostate and other cancers propelled us to investigate isoform specific functions in prostate cancer (PCa). One hundred and twenty laser capture micro-dissection matched normal prostate and prostate tumor tissues were analyzed for correlations between quantitative expression of PMEPA1 isoforms and clinical outcomes with Q-RT-PCR, and further validated with a The Cancer Genome Atlas (TCGA) RNA-Seq dataset of 499 PCa. Cell proliferation was assessed with cell counting, plating efficiency and soft agar assay in androgen responsive LNCaP and TGF-β responsive PC3 cells. TGF-β signaling was measured by SMAD dual-luciferase reporter assay. Higher PMEPA1-a mRNA levels indicated biochemical recurrence (p = 0.0183) and lower PMEPA1-b expression associated with metastasis (p = 0.0173). Further, lower PMEPA1-b and a higher ratio of PMEPA1-a vs. -b were correlated to higher Gleason scores and lower progression free survival rate (p < 0.01). TGF-β-responsive PMEPA1-a promoted PCa cell growth, and androgen-responsive PMEPA1-b inhibited cancer cell proliferation. PMEPA1 isoforms -a and -b were shown to be promising candidate biomarkers indicating PCa aggressiveness including earlier biochemical relapse and lower disease specific life expectancy via interrupting androgen/TGF-β signaling.

scholarly journals Five lncRNAs Associated With Prostate Cancer Prognosis Identified by Coexpression Network Analysis

2020 ◽  
Vol 19 ◽  
pp. 153303382096357
Author(s):  
Xiaoyong Gong ◽  
Bobin Ning
Keyword(s):  
Prostate Cancer ◽  
Network Analysis ◽  
Cox Regression ◽  
Characteristic Curve ◽  
Interaction Network ◽  
The Cancer Genome Atlas ◽  
Cancer Prognosis ◽  
Normal Prostate ◽  
Cox Regression Analysis ◽  
Incidence And Mortality

Prostate cancer (PCa) is a highly malignant tumor, with increasing incidence and mortality rates worldwide. The aim of this study was to identify the prognostic lncRNAs and construct an lncRNA signature for PCa diagnosis by the interaction network between lncRNAs and protein-coding genes (PCGs). The differentially expressed lncRNAs (DElncRNAs) and PCGs (DEPCGs) between PCa and normal prostate tissues were screened from The Cancer Genome Atlas (TCGA) database. The DEPCGs were functionally annotated in terms of the enriched pathways. Weighted gene co-expression network analysis (WGCNA) of 104 PCa samples identified 15 co-expression modules, of which the Turquoise module was negatively correlated with cancer and included 5 key lncRNAs and 47 PCGs. KEGG pathway analyses of the core 47 PCGs showed significant enrichment in classic PCa-related pathways, and overlapped with the enriched pathways of the DEPCGs. LINC00857, LINC00900, LINC00908, LINC00900, SNHG3 and FENDRR were significantly associated with the survival of PCa and have not been reported previously. Finally, Multivariable Cox regression analysis was used to establish a prognostic risk formula, and the patients were accordingly stratified into the low- and high-risk groups. The latter had significantly worse OS compared to the low-risk group (P < 0.01), and the area under the receiver operating characteristic curve (ROC) of 14-year OS was 0.829. The accuracy of our prediction model was determined by calculating the corresponding concordance index (C-index) and risk curves. In conclusion, we established a 5-lncRNA prognostic signature that provides insights into the biological and clinical relevance of lncRNAs in PCa.

Identification of the PCA29 gene signature as a predictor in prostate cancer

2019 ◽  
Vol 17 (03) ◽  
pp. 1940006 ◽  
Author(s):  
Jung-Yu Lee ◽  
Si-Yu Lin ◽  
Chun-Yu Lin ◽  
Yi-Huan Chuang ◽  
Sing-Han Huang ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Quantitative Polymerase Chain Reaction ◽  
Gene Signature ◽  
Gene Expression Omnibus ◽  
The Cancer Genome Atlas ◽  
Gene Expressions ◽  
Scoring Method ◽  
Clinical Issue ◽  
Targeting Therapy ◽  
Human Protein Atlas

Prostate cancer (PCa) is the second leading cause of cancer death among men worldwide. About 70% of PCa patients were diagnosed at later stage, and metastasis has been observed. Additionally, the cure rate of PCa closely relies on the early diagnosis with biomarkers. The identification of biomarkers for diagnosis and prognosis is an urgent clinical issue for PCa. Here, we developed a novel scoring strategy, including cluster score (CS) and predicting score (PS), to identify 29 PCa genes (called PCa29) for early diagnostic biomarkers from two datasets in Gene Expression Omnibus. The result indicates that PCa29 can discriminate between normal and tumor tissues and are specific for prostate cancer. To validate PCa29, we found that 97% of PCa29 were consistently significant with these gene expressions in The Cancer Genome Atlas; furthermore, [Formula: see text]70% of PCa29 are consensus to the protein expression in The Human Protein Atlas. Finally, we examined 10 genes in PCa29 on three PCa cell lines by real-time quantitative polymerase chain reaction. The experimental results show that the trend of the differential PCa29 expression is consistent with the analyzed results from our novel scoring method. We believe that our method is useful and PCa29 are potential biomarkers that provide the clues to develop targeting therapy for PCa.

Kallikreins are involved in an miRNA network that contributes to prostate cancer progression

2014 ◽  
Vol 395 (9) ◽  
pp. 991-1001 ◽  
Author(s):  
Sara Samaan ◽  
Zsuzsanna Lichner ◽  
Qiang Ding ◽  
Carol Saleh ◽  
Joseph Samuel ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Cancer Progression ◽  
Integrated Circuit ◽  
Serine Proteases ◽  
Target Genes ◽  
Target Prediction ◽  
Preliminary Evidence ◽  
Differentially Expressed ◽  
Prostate Cancer Progression ◽  
Microsatellite Repeats

Abstract MicroRNAs (miRNAs) are short RNA nucleotides that negatively regulate their target genes. They are differentially expressed in prostate cancer. Kallikreins are genes that encode serine proteases and are dysregulated in cancer. We elucidated a miRNA-kallikrein network that can be involved in prostate cancer progression. Target prediction identified 23 miRNAs that are dysregulated between high and low risk biochemical failure and are predicted to target five kallikreins linked to prostate cancer; KLK2, KLK3, KLK4, KLK14 and KLK15. We also identified 14 miRNAs that are differentially expressed between Gleason grades and are predicted to target these kallikreins. This demonstrates that kallikreins are downstream effectors through which miRNAs influence tumor progression. We show, through in-silico and experimental analysis, that miR-378/422a and its gene targets PIK3CG, GRB2, AKT3, KLK4 and KLK14 form an integrated circuit in prostate cancer. Our analysis shows that a minisatellite sequence in the kallikrein locus consists of a number of microsatellite repeats that represent predicted miRNA response elements. A number of kallikrein and non-kallikrein prostate cancer-related genes share these microsatellite repeats. We validated some of these interactions in prostate cancer cell lines. Finally, we provide preliminary evidence on the presence of a miRNA-mediated cross-talk between kallikreins, including a kallikrein pseudogene.

Sign in / Sign up

Export Citation Format

Share Document