Pediatric CNS tumor classification by DNA-methylation dataset
Ontology highlight
ABSTRACT:
PROVIDER: EGAD00010002599 | EGA |
REPOSITORIES: EGA
Similar Datasets
Project description:In this study nanopore sequencing was applied to obtain sparse DNA methylation profiles from pediatric CNS tumor samples. A neural network was used to classify the tumor based on the obtained methylation profile.
2023-08-08 | GSE237874 | GEO
Project description:In this study nanopore sequencing was applied to obtain sparse DNA methylation profiles from pediatric CNS tumor samples. A neural network was used to classify the tumor based on the obtained methylation profile.
| EGAD00001011303 | EGA
Project description:Acute leukemias (AL) are aggressive blood cancers that require precise molecular classification and urgent treatment. However, standard-of-care diagnostic tests are time and resource intensive and do not capture the full spectrum of AL heterogeneity. Here, we developed a machine learning framework to classify AL using genome-wide DNA methylation profiling. We first assembled a large reference cohort (n=2,540 samples) and defined 38 distinct methylation classes across AL lineages and age groups. A subset of methylation classes mirrored established AL categories defined by genetics, such as AML with CBFB::MYH11 and B-ALL with ETV6::RUNX1, while others revealed epigenetic heterogeneity not captured by standard-of-care testing alone, including four methylation classes associated with NPM1-mutant AML (HOXA/B-activated) and four with KMT2A-rearranged AML (HOXA-activated). Using this reference, we next developed a deep neural network model (MARLIN) for methylation-based AL classification from extremely sparse data, and applied this for rapid analysis of clinical samples profiled by nanopore sequencing. In a retrospective AL cohort, MARLIN-based classifications were concordant with standard-of-care diagnoses in 25/26 (96.2%) cases with high-confidence prediction scores, including samples of diverse lineages and molecular subtypes. In five additional patients presenting with suspected AL, we performed nanopore sequencing and MARLIN classification of clinical blood and bone marrow samples in real-time, typically achieving an accurate methylation class prediction in less than two hours from sample receipt and less than one hour of sequencing time. In summary, we present a DNA methylation-based machine learning framework for rapid AL classification in the clinic that can complement and enhance standard-of-care diagnostics.
2025-06-16 | GSE280090 | GEO
Project description:The primary treatment of CNS tumors starts with a neurosurgical resection in order to obtain tumor tissue for diagnosis and to reduce tumor load and mass effect. The neurosurgeon has to decide between radical resection versus a more conservative strategy to prevent surgical morbidity. The prognostic impact of a radical resection varies between tumor types. However due to a lack of pre-operative tissue-based diagnostics, limited knowledge of the precise tumor type is available at the time of surgery. Current standard practice includes preoperative imaging and intraoperative histological analysis, but these are not always conclusive. After surgery, histopathological and molecular tests are performed to diagnose the precise tumor type. The results may indicate that an additional surgery is needed or that the initial surgery could have been less radical. Using rapid Nanopore sequencing, a sparse methylation profile can be directly obtained during surgery, making it ideally suited to enable intraoperative diagnostics. We developed a state-of-the-art neural-network approach called Sturgeon, to deliver trained models that are lightweight and universally applicable across patients and sequencing depths. We demonstrate our method to be accurate and fast enough to provide a correct diagnosis with as little as 20 to 40 minutes of sequencing data in 45 out of 49 pediatric samples, and inconclusive results in the other four. In four intraoperative cases we achieved a turnaround time of 60-90 minutes from sample biopsy to result; well in time to impact surgical decision making. We conclude that machine-learned diagnosis based on intraoperative sequencing can assist neurosurgical decision making, allowing neurological comorbidity to be avoided or preventing additional surgeries.
| EGAS00001007475 | EGA
Project description:DNA methylation-based classification of brain tumors has emerged as a powerful and indispensable diagnostic technique. Initial implementations have used methylation microarrays for data generation, but different sequencing approaches are increasingly used. Most current classifiers, however, rely on a fixed methylation feature space, rendering them incompatible with other platforms, especially different flavors of DNA sequencing. Here, we describe crossNN, a neural network-based machine learning framework which can accurately classify tumor entities using DNA methylation profiles obtained from different platforms and with different epigenome coverage and sequencing depth. It outperforms other deep and conventional machine learning models with respect to diagnostic accuracy and computational requirements while still being fully explainable. We use crossNN to train a pan-cancer classifier to discriminate more than 170 tumor types across all organ sites. Validation in an independent cohort of >5,000 tumors profiled using different microarray and sequencing platforms, including low-pass nanopore and targeted bisulfite sequencing, demonstrates the robustness and scalability of the model with 99.1% and 97.8% precision for brain tumor and pan-cancer model, respectively.
2025-02-24 | GSE289137 | GEO
Project description:DNA methylation-based classification of brain tumors has emerged as a powerful and indispensable diagnostic technique. Initial implementations have used methylation microarrays for data generation, but different sequencing approaches are increasingly used. Most current classifiers, however, rely on a fixed methylation feature space, rendering them incompatible with other platforms, especially different flavors of DNA sequencing. Here, we describe crossNN, a neural network-based machine learning framework which can accurately classify tumor entities using DNA methylation profiles obtained from different platforms and with different epigenome coverage and sequencing depth. It outperforms other deep and conventional machine learning models with respect to diagnostic accuracy and computational requirements while still being fully explainable. We use crossNN to train a pan-cancer classifier to discriminate more than 170 tumor types across all organ sites. Validation in an independent cohort of >5,000 tumors profiled using different microarray and sequencing platforms, including low-pass nanopore and targeted bisulfite sequencing, demonstrates the robustness and scalability of the model with 99.1% and 97.8% precision for brain tumor and pan-cancer model, respectively.
2025-02-24 | GSE289246 | GEO
Project description:Cerebrospinal fluid (CSF) liquid biopsies serve as a rich source of tumor-derived cell-free DNA (cfDNA) for evaluating patients with central nervous system (CNS) tumors. However, challenges stemming from trace cfDNA yields and low mutational burden have hindered sensitivity, whereas first-generation clinical assays have relied on genetic alterations as biomarkers. Leveraging the diagnostic utility of DNA methylation classification in CNS tumors, we developed M-PACT (Methylation-based Predictive Algorithm for CNS Tumors), a robust deep neural network that accurately classifies tumors from sub-nanogram input cfDNA methylomes acquired through enzymatic methylation sequencing. In addition to tumor classification, this workflow enables methylation-based cellular deconvolution and sensitive copy number variation (CNV) detection. We benchmark our methodology in pediatric CNS embryonal tumors and further demonstrate accurate classification of intra-operative CSF, balanced tumor genomes, and secondary malignancies. Altogether, we provide a blueprint for CNS tumor classification from low input cfDNA methylomes, motivating prospective validation for future clinical implementation.
2025-11-17 | GSE292312 | GEO
Project description:Ultra-fast deep-learned pediatric CNS tumor classification
| PRJNA996977 | ENA
Project description: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.
2024-10-10 | GSE276299 | GEO
Project description:DNA methylation profiling is commonly utilized in tumor diagnosis and has shown diagnostic specificity in many contexts. We recently noted that several CNS tumor classifiers used for neuropathology clinical diagnostics fail to appropriately classify pediatric cerebellar cortex tissue as non-neoplastic. Our study illustrates a gap in classifier accuracy that is most evident with samples from very young patients (<6 months). In this setting,classifiers typically assign medulloblastoma, SHH-activated, and occasionally show high-confidence scores. Neuropathologists practicing in pediatric settings should be aware of the histologic appearance of immature cerebellar tissue, and exercise caution when performing methylation profiling of samples of uncertain neoplastic status for patients in early infancy.
2025-07-02 | GSE299377 | GEO