A machine learning breast cancer prediction model based on a panel from circulating exosomal miRNAs
Ontology highlight
ABSTRACT:
ORGANISM(S): Homo sapiens
PROVIDER: GSE197020 | GEO | 2022/02/21
REPOSITORIES: GEO
Similar Datasets
Project description:Background: Circular RNA (circRNA) has been reported to be involved in the pathogenesis of cardiovascular disease (CVD), however, it is unclear whether circRNA carried by exosomes (exos) can be used as biomarkers for chronic coronary syndrome (CCS). This study aimed to investigate the expression pattern of exosomal circRNAs in the plasma of CCS patients, and to screen exo-circRNAs that can act as biomarkers for the diagnosis of CCS. Methods: High-throughput sequencing technology was carried out in the plasma exosomal RNA of 15 CCS patients and 15 non-cardiac chest pain patients (NCCP, control group) (sequencing cohort) to screen for differentially expressed circRNAs. Selected differentially expressed exo-circRNAs were further verified by real time polymerase chain reaction (RT-PCR) in a small-sample cohort (derivation cohort) and a large-sample cohort (validation cohort). Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic value of exo-circRNAs for CCS patients. Results: By high-throughput sequencing of circRNAs in 15 CCS patients and 15 NCCP patients, 276 circRNAs differentially expressed in plasma exosomes of CCS patients were screened by using |log2(Fold change) |>1 and q-value<0.001, and 103 were up-regulated and 173 were down-regulated. Among the 103 up-regulated circRNAs, 5 circRNAs with high expression levels and significantly increased in plasma exosomes of CCS patients were selected for validation. Twice RT-PCR validation demonstrated that exo-hsa_circ_0075269 and exo-hsa_circ_0000284 were significantly up-regulated in patients with CCS (P<0.001). Circulating exo-hsa_circ_0075269 and exo-hsa_circ_0000284 yielded the area under the curve (AUC) values of 0.761 (P<0.001, 95%CI=0.669, 0.852) and 0.623 (P=0.015, 95%CI=0.522, 0.724) for CCS, respectively by ROC curve analysis. Moreover, AUC of hsa_circ_0075269 combined with hsa_circ_0000284 for diagnosing CCS was 0.741 (P<0.001, 95%CI=0.667, 0.850). Conclusion: The expression profile of circRNA in plasma exosomes of patients with CCS was significantly different from that of the control group. Plasma exo-hsa_circ_0075269 and exo-hsa_circ_0000284 have the potential to be new biomarkers for CCS diagnosis.
2023-11-09 | GSE237983 | GEO
Project description:MiRNA profiling from whole blood (EDTA) was used to identify a miRNA profile applicable for early stage breast cancer detection In the study presented here, a consecutively collected, well-defined cohort of 48 patients with early stage breast cancer (invasive ductal carcinoma) at diagnosis was compared with a cohort of 57 healthy individuals for known homo sapiens miRNAs (mirbase 15).
2012-02-02 | E-GEOD-31309 | biostudies-arrayexpress
Project description:To investigate miRNA expression profiles between gastric cancers plasma and healthy controls plasma,screening the microRNAs that can serve as plasma biomarkers for the early diagnosis of gastric cancer by miRNA array.
2016-09-13 | GSE86822 | GEO
Project description:Neuroblastoma (NB) is the most common extracranial solid tumor during infancy, causing up to 10% of mortality in children. NB shows notable heterogeneity with regard to histology and clinical behavior, ranging from low-risk localized tumors (LR-NB) to high-risk disease (HR-NB), characterized by aggressive metastatic phenotype, resistance to treatment, and fatal relapse occurrence. Risk stratification demands high accuracy, as it will determine the therapeutic treatment. However, the current therapeutic stratification, based on clinical and molecular risk factors, does not allow to discriminate among patients with similar clinical-pathological parameters who receive the same treatment despite showing markedly different clinical courses. The need of novel diagnostic molecular tools in oncology has led to increasing interest in liquid biopsies as a source of biomarkers, as they provide a minimally invasive method. Body fluids are a source of exosomes, nanosized extracellular vesicles that can drive tumor growth and chemoresistance. We previously identified exosomal microRNAs (exo-miRNA) indicative of HR-NB patient sensitivity/resistance to chemotherapy. As exosomes from cancer patients carry proteins that reflect the surface and cytoplasm content of parental cancer cells, including immunosuppressive molecules known to be associated with cancer progression and/or response to therapy, we extended our study to the exosomal proteins (Exo-prot).The purpose of the study is to identify Exo-prot that (i) are specifically expressed in NB patients and (ii) are associated with tumor phenotype and disease stage, in order to improve risk stratification and refine diagnosis.We isolated exosomes from plasma specimens of 24 HR-NB patients and 24 LR-NB patients at diagnosis and of 24 age-matched healthy controls (CTRL). Exo-prot expression was measured by liquid chromatography-mass spectrometry. Missing value distribution and differential expression analysis (performed with two different imputation methods) were applied to identify the most relevant Exo-prot. ROC analysis assessed the diagnostic value of the identified Exo-prot. We demonstrated that NB patients have a different Exo-prot expression profile compared to CTRL. The deregulated Exo-prot in NB specimens act mainly in tumor-associated pathways and build a strictly connected network. Furthermore, HR-NB patients show a different Exo-miR expression profile compared to LR-NB subjects, with the modulation of molecules involved in cell migration, proliferation and metastasis. Importantly, we show that NCAM, NCL, LUM and VASP have a diagnostic value in discriminating NB patients from CTRL; while MYH9, FN1, CALR, AKAP12 and, with a lower performance, LTBP1 can differentiate HR-NB and LR-NB patients with high accuracy. We demonstrated that Exo-prot deregulated in NB vs CTRL subjects and in HR-NB vs LR-NB patients contribute to NB tumor development and to the aggressive metastatic NB phenotype, respectively. We identified Exo-prot with significant diagnostic value for NB patients and able to efficiently identify the HR-NB subset of patients, which can be employed to improve risk stratification. Our results highlight the applicability of Exo-prot evaluation for integrating NB diagnosis and risk assessment.
2024-01-26 | PXD042422 | Pride
Project description:Study investigating the role of miRNAs in breast cancer detection miRNA extracted from plasma on 20 breast cancer patients, 20 controls, 20 post-resection breast cancer patients and 10 lung/colorectal cancer patients, miRNA quanitification using Illumina microarray
2012-10-13 | E-GEOD-41526 | biostudies-arrayexpress
Project description:Early-onset preeclampsia (EOPE) is a severe pregnancy complication associated with significant maternal and fetal morbidity and mortality. Currently, there is a critical need for accurate, non-invasive biomarkers to facilitate early diagnosis and effective management of EOPE. In this study, we aimed to investigate the transcriptional alterations and non-invasive biomarker potential of peripheral blood microRNAs in patients with EOPE. Through our research, we successfully identified two reliable plasma miRNA biomarkers and proposed a circulating two-miRNA panel for the non-invasive early detection of EOPE. Additionally, we independently validated our findings in different patient cohorts using various technological platforms.
2023-06-14 | GSE234611 | GEO
Project description:Low fertility remains a leading cause of poor productivity in dairy cattle. In this context, there is significant interest in developing novel tools for accurate early diagnosis of pregnancy. MicroRNAs (miRNAs) are short RNA molecules which are critically involved in regulating gene expression during both health and disease. MiRNAs have been shown to regulate ovarian function, uterine receptivity, embryonic development and placental function. Circulating miRNAs can provide useful biomarkers of tissue function and disease; importantly, differential miRNA profiles have been linked to pregnancy and preeclampsia in humans. This study sought to establish the potential of circulating miRNAs as biomarkers of early pregnancy in cattle. We applied Illumina small-RNA sequencing to profile miRNAs in plasma samples collected from eight non-pregnant heifers on Days 0, 8 and 16 of the oestrous cycle and 11 heifers on Days 16 and 24 of pregnancy. We sequenced a total of 46 samples and generated 9.2 million miRNA reads per sample. There were no differences in miRNA read abundance between any of the pregnant and non-pregnant time-points (FDR > 0.1). As a complementary approach, we analysed sample pools (3-4 samples/pool) corresponding to Days 0, 8 and 16 of the oestrous cycle and Day 24 of pregnancy (n = 3 pools/group) using Qiagen PCR arrays. A total of 16 miRNAs were differentially expressed (FDR < 0.1) in plasma between pregnant and non-pregnant animals. RT-qPCR validation using the same plasma samples confirmed that miR-26a was differentially upregulated on Day 16 pregnant relative to non-pregnant heifers (1.7-fold; P = 0.043), whereas miR-1249 tended to be upregulated in Day 16 pregnant heifers (1.6-fold; P = 0.081). Further validation in an independent group of heifers confirmed an increase in plasma miR-26a levels during early pregnancy, which was significant only on Day 24 (2.0-fold; P = 0.027). Through genome-wide analyses we have successfully profiled plasma miRNA populations associated with early pregnancy in cattle. We have identified miR-26a as a potential circulating biomarker of early pregnancy.
2016-08-03 | GSE85090 | GEO
Project description:The paper "Metabolomic Machine Learning Predictor for Diagnosis and Prognosis of Gastric Cancer" addresses the need for non-invasive diagnostic tools for gastric cancer (GC). Traditional methods like endoscopy are invasive and expensive. The authors conducted a targeted metabolomics analysis of 702 plasma samples to develop machine learning models for GC diagnosis and prognosis. The diagnostic model, using 10 metabolites, achieved a sensitivity of 0.905, outperforming conventional protein marker-based methods. The prognostic model effectively stratified patients into risk groups, surpassing traditional clinical models.
I have successfully reproduced the diagnosis model from the paper. This machine learning-based system differentiates GC patients from non-GC controls using metabolomics data from plasma samples analyzed by liquid chromatography-mass spectrometry (LC-MS). The model focuses on 10 metabolites, including succinate, uridine, lactate, and serotonin. Employing LASSO regression and a random forest classifier, the model achieved an AUROC of 0.967, with a sensitivity of 0.854 and specificity of 0.926. This model significantly outperforms traditional diagnostic methods and underscores the potential of integrating machine learning with metabolomics for early GC detection and treatment.
2024-07-23 | MODEL2407230001 | BioModels
Project description:To achieve the best outcomes, breast cancer necessitates robust strategies for early detection. However, reliable blood-based tests for identifying early-stage disease remains elusive. Here we have employed plasma metabolomics and machine learning techniques to establish a non-invasive metabolic approach for early detection of breast cancer.
2023-11-06 | MSV000093325 | MassIVE
Project description:To explore the possibility of miRNA(s) contributing to the cardioprotection induced by plasma exosomes at the late phase of RIPC, we performed a miRNA profiling assay (763 rat miRNAs) comparing the differences between RIPC-exo and Control-exo using Illumina HiSeq 2500 high-throughput sequencing.
2021-12-23 | GSE182675 | GEO