GEO Logo
   NCBI > GEO > Accession DisplayHelp Not logged in | LoginHelp
GEO help: Mouse over screen elements for information.
Series GSE90496 Query DataSets for GSE90496
Status Public on Jan 14, 2018
Title DNA methylation-based classification of human central nervous system tumors [reference set]
Organism Homo sapiens
Experiment type Methylation profiling by genome tiling array
Summary Background: Modern neuropathology is challenged by an increasing number of clinically-relevant CNS tumor subgroups that require assessment of a multitude of molecular markers for classification, as well a highly trained medical staff. Failure to meet this challenge leads to tumor misclassification, which can have severe consequences for affected patients. Methods: We compiled a cohort of genome-wide DNA methylation profiles of 2,682 tumors from 82 histologically and/or molecularly distinct CNS tumor classes across all ages and histologies that served as reference for a Random Forest-based diagnostic classifier. This classifier was used to prospectively investigate a further 1,104 CNS tumor samples in order to determine its clinical utility. Results: The classifier was able to reliably assign tumor samples to a given diagnostic category with a misclassification rate of less than 2%. The system functioned robustly across laboratories and using different DNA methylation profiling techniques. Prospective application to clinical samples resulted in a reclassification of 12% of tumors compared with standard practice alone. A further 12% could not be classified by methylation profiling – this subset was highly enriched for unusual syndrome-associated tumors and likely novel entities. Conclusion: This study represents a proof-of-concept for the application of machine learning approaches in molecular diagnostics using a single, easy-to-use assay. The reference cohort and Random Forest-based classifier are available online as a valuable community tool for improving precision in brain tumor diagnostics. We expect that approaches similar to the one presented herein will rapidly restructure diagnostic practice in neurooncology and across tumor pathology.
Overall design The Illumina Infinium HumanMethylation450 array (Illumina, San Diego, USA) data from n=2801 patients were used to obtain genome-wide assessment of DNA methylation, according to the manufacturer’s instructions at the Genomics and Proteomics Core Facility of the German Cancer Research Center (DKFZ) as described previously (Sturm et al. Cancer Cell. 2012 Oct 16;22(4):425-37)
Contributor(s) Capper D, Jones DW, Sill M, Hovestadt V, von Deimling A, Pfister SM
Citation(s) 29539639, 36964296
Submission date Nov 25, 2016
Last update date Jan 23, 2024
Contact name Martin Sill
Organization name DKFZ
Department Division of Pediatric Neurooncology
Street address Im Neuenheimer Feld 280
City Heidelberg
ZIP/Postal code 69120
Country Germany
Platforms (1)
GPL13534 Illumina HumanMethylation450 BeadChip (HumanMethylation450_15017482)
Samples (2801)
GSM2402854 GBM, G34, sample 1 [reference set]
GSM2402855 DMG, K27, sample 2 [reference set]
GSM2402856 DMG, K27, sample 3 [reference set]
This SubSeries is part of SuperSeries:
GSE109381 DNA methylation-based classification of human central nervous system tumors
BioProject PRJNA354936

Download family Format
SOFT formatted family file(s) SOFTHelp
MINiML formatted family file(s) MINiMLHelp
Series Matrix File(s) TXTHelp

Supplementary file Size Download File type/resource
GSE90496_RAW.tar 22.7 Gb (http)(custom) TAR (of IDAT)
GSE90496_beta.txt.gz 9.2 Gb (ftp)(http) TXT
GSE90496_methylationclassdescription.xlsx 17.2 Kb (ftp)(http) XLSX
Processed data are available on Series record

| NLM | NIH | GEO Help | Disclaimer | Accessibility |
NCBI Home NCBI Search NCBI SiteMap