Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Comparison of DNA Methylation Based Classification Models for Precision Diagnostics of Central Nervous System Tumors

ABSTRACT: As part of the advancement in therapeutic decision-making for brain tumor patients at St. Jude Children’s Research Hospital (SJCRH), we developed three robust classifiers, a deep learning neural network (NN), k-nearest neighbor (kNN), and random forest (RF), trained on a reference series DNA-methylation profiles to classify central nervous system (CNS) tumor types. The models’ performance was rigorously validated against 2,054 samples from two independent cohorts. In addition to classic metrics of model performance, we compared the robustness of the three models to reduced tumor purity, a critical consideration in the clinical utility of such classifiers. Our findings revealed that the NN model exhibited the highest accuracy and maintained a balance between precision and recall. The NN model was the most resistant to drops in performance associated with a reduction in tumor purity, showing good performance until the purity fell below 50%. Through rigorous validation, our study emphasizes the potential of DNA-methylation-based deep learning methods to improve precision medicine for brain tumor classification in the clinical setting.

ORGANISM(S): Homo sapiens

PROVIDER: GSE276299 | GEO | 2024/10/10

REPOSITORIES: GEO

ACCESS DATA

Json Xml

Dataset's files

Source:

			Action	DRS
		Other

Items per page:

1 - 1 of 1

Similar Datasets

Machine Learning Models Predict the Primary Sites of Head and Neck Squamous Cell Carcinoma Metastases Based on DNA Methylation

Project description:In head and neck squamous cell cancers (HNSCs) that present as metastases with an unknown primary (HNSC-CUPs), the identification of a primary tumor improves therapy options and increases patient survival. However, the currently available diagnostic methods are laborious and do not offer a sufficient detection rate. Predictive machine learning models based on DNA methylation profiles have recently emerged as a promising technique for tumor classification. We applied this technique to HNSC to develop a tool that can improve the diagnostic workup for HNSC-CUPs. On a reference cohort of 405 primary HNSC samples, we developed four classifiers based on different machine learning models (random forest (RF), neural network (NN), elastic net penalized logistic regression (LOGREG), support vector machine (SVM)) that predict the primary site of HNSC tumors from their DNA methylation profile. The classifiers achieved high classification accuracies (RF=83%, NN=88%, LOGREG=SVM=89%) on an independent cohort of 64 HNSC metastases. Further, the NN, LOGREG, and SVM models significantly outperformed p16 status as a marker for an origin in the oropharynx. In conclusion, the DNA methylation profiles of HNSC metastases are characteristic for their primary sites and the classifiers developed in this study, which are made available to the scientific community, can provide valuable information to guide the diagnostic workup of HNSC-CUP.

2021-12-08 | GSE171994 | GEO

Methylation profiling of plasma cell-free DNA in pediatric brain tumor patients

Project description:Using cell-free methylated DNA immunoprecipitation and high-throughput sequencing (cfMeDIP-seq), we analyzed methylation profiles in 93 plasma samples from 77 pediatric brain tumor patients and 16 non-neoplastic patients. Binomial GLMnet classifiers of tumor and tumor subtypes were built in training sets, and performance was evaluated in test sets. The methylation profiles from plasma cfMeDIP-seq discriminated tumor from non-tumor patients with 0.83 accuracy (precision = 0.93, sensitivity = 0.86, and specificity = 0.67). Among major tumor subtypes vs. nontumors, circumscribed astrocytic glioma showed an accuracy of 0.86 (precision = 0.88, sensitivity = 0.88, and specificity = 0.83), and glioneuronal tumors had an accuracy of 0.83 (precision, sensitivity, and specificity = 0.83). Circumscribed astrocytic glioma and glioneuronal tumors could be discriminated from other tumor subtypes with 0.79 and 0.82 accuracy, respectively.

2024-09-30 | GSE275598 | GEO

A hybrid signal/sequence model to detect RNA polymerase activity in ATAC-seq

Project description:We explore different uses of machine learning classifiers, including neural networks, to combine the “signal” from ATAC-seq with its underlying genome sequence in order to classify ATAC-seq peaks on the presence or absence of transcription. We show how a hybrid signal/sequence representation, classified using recurrent neural networks (RNNs), yields the best performance across different cell types.

2020-04-30 | GSE130205 | GEO

Comprehensive RNA-Seq Dataset for Pan-Cancer Tumor Classification

Project description:The following consists of an RNA-Seq cohort comprised of samples from patients diagnosed with various tumor types across the pan-cancer spectrum. Collected at diagnosis, these samples serve as a resource for tumor classification using machine learning classifiers. Process count tables are provided. This cohort can assist researchers in exploring molecular signatures and facilitate accurate classification across diverse malignancies.

2024-06-30 | E-MTAB-14038 | biostudies-arrayexpress

Validation of Whole Genome Methylation Profiling Classifier for Central Nervous System Tumors

Project description:The 2021 WHO Classification of Tumors of the Central Nervous System includes several tumor types and subtypes for which the diagnosis is at least partially reliant on utilization of whole genome methylation profiling. The current approach to array DNA methylation profiling utilizes a reference library of tumor DNA methylation data, and a machine learning-based tumor classifier. This approach was pioneered and popularized by the German Cancer Research Network (DKFZ) and University Hospital Heidelberg. This research group has kindly made their classifier for central nervous system tumors freely available as a research tool via a web-based portal. However, this classifier is not maintained in a clinical testing environment. Therefore, we validated our own DNA methylation-based classifier of central nervous system tumors. We validated our classifier using the same training and validation datasets as the DKFZ group. In addition, we performed a validation of samples tested in our own laboratory and compared the performance of both classifiers. Using the validation data set, our classifier’s performance showed high concordance (92%) and comparable accuracy (specificity 94.0% v. 84.9% for DKFZ, sensitivity 88.6% v. 94.7% for DKFZ). Receiver operator curve showed areas under the curve of 0.964 v. 0.966 for NM and DKFZ classifiers, respectively. Our classifier performed comparably well with samples tested in our own laboratory and is currently offered for clinical testing.

2022-03-19 | GSE198855 | GEO

Patterson2022 - Tumour mutation data driven Random Forest model to predict immune checkpoint inhibitor therapy benefit in metastatic melanoma

Project description:A Random Forest model is developed to incorporate tumor mutation data within the context of the biological process known as leukocyte proliferation regulation. This model aims to predict a patient's response to anti-PD1 treatment. The authors conducted experiments using four different types of classifiers: Random Forest, Gradient Boosting, Feed Forward Neural Network, and Long Short-Term Memory (LSTM) recurrent neural network. Among these classifiers, the Random Forest algorithm yielded the best predictive performance when modeling gene mutation data associated with the 'leukocyte proliferation regulation' biological process. Hence, this curated version of the model focuses on the Random Forest model trained specifically on the 'Leukocyte Proliferation Regulation' process. In this model, a value of '0' is assigned to NonResponders, while a value of '1' is assigned to Responders. Please note that to obtain predictions, users should provide mutation data containing only the genes corresponding to the 'GO_REGULATION_OF_LEUKOCYTE_PROLIFERATION' process keyword, as specified in the 'GO_test_genes_dict_intersection' dictionary.

2023-07-03 | BIOMD0000001073 | BioModels

Project description:Tumor Purity in Preclinical Mouse Tumor Models

| PRJNA800199 | ENA

Project description:Performance and precision of ddRAD

| PRJNA387279 | ENA

Genome profiling of free circulating tumor DNA by oncoscan and comparison to profiles in tumor biopsies evaluated by cytoscan

Project description:Genomic profiles of tumor DNA obtained from biopsies were systematically compared to corresponding samples of circulating cell free tumor DNA. The aim was to investigate correspondance between numerical, segmental and focal aberrations, as well as precision in detection and resolution of oncoscan array performance on cfDNA.

2016-11-30 | GSE85937 | GEO

Proteogenomic Analysis of Enriched Tumor Epithelium Identifies Prognostic Signatures and an Increased Dependency of Homologous Recombination Proficient Cells on BMI1 in High Grade Serous Ovarian Cancer

Project description:Although several proteogenomic analyses of high-grade serous ovarian cancer (HGSOC) have been conducted, these analyses are often dominated by tissues with high levels of tumor cell nuclei (purity), despite low purity being associated with worse outcome. We performed deep proteogenomic profiling of bulk tumor (BT) and laser microdissection (LMD) enriched tumor (ET) and stromal (ST) cell populations from 70 HGSOC patient tumors spanning a broad spectrum of purity. We identified a cluster of patients with longer progression free survival associated with increased immune signatures and validated proteins correlating with tumor infiltrating lymphocytes (TILs) in 65 tumors collected from an independent cohort of 12 HGSOC patients, as well as with overall survival in an additional cohort of 126 HGSOC patients. We identified that homologous recombination deficient (HRD) tumors express transcriptomic and proteomic pathways associated with metabolism and oxidative phosphorylation that we validated in independent patient cohorts, including 69 HRD-positive HGSOC tumors. In summary, our proteogenomic analysis provides important new clinically relevant insights into HGSOC tumor cell populations, and have uncovered prognostic proteogenomic alterations correlating with TILs, low tumor purity, as well as expression alterations associated with HRD status and immune infiltration.

2024-03-15 | PXD045417 | Pride

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data