Dataset Information

DNA methylation-based classification of central nervous system tumours.

ABSTRACT: Accurate pathological diagnosis is crucial for optimal management of patients with cancer. For the approximately 100 known tumour types of the central nervous system, standardization of the diagnostic process has been shown to be particularly challenging-with substantial inter-observer variability in the histopathological diagnosis of many tumour types. Here we present a comprehensive approach for the DNA methylation-based classification of central nervous system tumours across all entities and age groups, and demonstrate its application in a routine diagnostic setting. We show that the availability of this method may have a substantial impact on diagnostic precision compared to standard methods, resulting in a change of diagnosis in up to 12% of prospective cases. For broader accessibility, we have designed a free online classifier tool, the use of which does not require any additional onsite data processing. Our results provide a blueprint for the generation of machine-learning-based tumour classifiers across other cancer entities, with the potential to fundamentally transform tumour pathology.

SUBMITTER: Capper D

PROVIDER: S-EPMC6093218 | biostudies-literature | 2018 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

DNA methylation-based classification of central nervous system tumours.

Capper David D Jones David T W DTW Sill Martin M Hovestadt Volker V Schrimpf Daniel D Sturm Dominik D Koelsche Christian C Sahm Felix F Chavez Lukas L Reuss David E DE Kratz Annekathrin A Wefers Annika K AK Huang Kristin K Pajtler Kristian W KW Schweizer Leonille L Stichel Damian D Olar Adriana A Engel Nils W NW Lindenberg Kerstin K Harter Patrick N PN Braczynski Anne K AK Plate Karl H KH Dohmen Hildegard H Garvalov Boyan K BK Coras Roland R Hölsken Annett A Hewer Ekkehard E Bewerunge-Hudler Melanie M Schick Matthias M Fischer Roger R Beschorner Rudi R Schittenhelm Jens J Staszewski Ori O Wani Khalida K Varlet Pascale P Pages Melanie M Temming Petra P Lohmann Dietmar D Selt Florian F Witt Hendrik H Milde Till T Witt Olaf O Aronica Eleonora E Giangaspero Felice F Rushing Elisabeth E Scheurlen Wolfram W Geisenberger Christoph C Rodriguez Fausto J FJ Becker Albert A Preusser Matthias M Haberler Christine C Bjerkvig Rolf R Cryan Jane J Farrell Michael M Deckert Martina M Hench Jürgen J Frank Stephan S Serrano Jonathan J Kannan Kasthuri K Tsirigos Aristotelis A Brück Wolfgang W Hofer Silvia S Brehmer Stefanie S Seiz-Rosenhagen Marcel M Hänggi Daniel D Hans Volkmar V Rozsnoki Stephanie S Hansford Jordan R JR Kohlhof Patricia P Kristensen Bjarne W BW Lechner Matt M Lopes Beatriz B Mawrin Christian C Ketter Ralf R Kulozik Andreas A Khatib Ziad Z Heppner Frank F Koch Arend A Jouvet Anne A Keohane Catherine C Mühleisen Helmut H Mueller Wolf W Pohl Ute U Prinz Marco M Benner Axel A Zapatka Marc M Gottardo Nicholas G NG Driever Pablo Hernáiz PH Kramm Christof M CM Müller Hermann L HL Rutkowski Stefan S von Hoff Katja K Frühwald Michael C MC Gnekow Astrid A Fleischhack Gudrun G Tippelt Stephan S Calaminus Gabriele G Monoranu Camelia-Maria CM Perry Arie A Jones Chris C Jacques Thomas S TS Radlwimmer Bernhard B Gessi Marco M Pietsch Torsten T Schramm Johannes J Schackert Gabriele G Westphal Manfred M Reifenberger Guido G Wesseling Pieter P Weller Michael M Collins Vincent Peter VP Blümcke Ingmar I Bendszus Martin M Debus Jürgen J Huang Annie A Jabado Nada N Northcott Paul A PA Paulus Werner W Gajjar Amar A Robinson Giles W GW Taylor Michael D MD Jaunmuktane Zane Z Ryzhova Marina M Platten Michael M Unterberg Andreas A Wick Wolfgang W Karajannis Matthias A MA Mittelbronn Michel M Acker Till T Hartmann Christian C Aldape Kenneth K Schüller Ulrich U Buslei Rolf R Lichter Peter P Kool Marcel M Herold-Mende Christel C Ellison David W DW Hasselblatt Martin M Snuderl Matija M Brandner Sebastian S Korshunov Andrey A von Deimling Andreas A Pfister Stefan M SM

Nature 20180314 7697

Accurate pathological diagnosis is crucial for optimal management of patients with cancer. For the approximately 100 known tumour types of the central nervous system, standardization of the diagnostic process has been shown to be particularly challenging-with substantial inter-observer variability in the histopathological diagnosis of many tumour types. Here we present a comprehensive approach for the DNA methylation-based classification of central nervous system tumours across all entities and ...[more]

PMID: 29539639

Similar Datasets

Project description:BackgroundPrecision medicine for cancer treatment relies on an accurate pathological diagnosis. The number of known tumor classes has increased rapidly, and reliance on traditional methods of histopathologic classification alone has become unfeasible. To help reduce variability, validation costs, and standardize the histopathological diagnostic process, supervised machine learning models using DNA-methylation data have been developed for tumor classification. These methods require large labeled training data sets to obtain clinically acceptable classification accuracy. While there is abundant unlabeled epigenetic data across multiple databases, labeling pathology data for machine learning models is time-consuming and resource-intensive, especially for rare tumor types. Semi-supervised learning (SSL) approaches have been used to maximize the utility of labeled and unlabeled data for classification tasks and are effectively applied in genomics. SSL methods have not yet been explored with epigenetic data nor demonstrated beneficial to central nervous system (CNS) tumor classification.ResultsThis paper explores the application of semi-supervised machine learning on methylation data to improve the accuracy of supervised learning models in classifying CNS tumors. We comprehensively evaluated 11 SSL methods and developed a novel combination approach that included a self-training with editing using support vector machine (SETRED-SVM) model and an L2-penalized, multinomial logistic regression model to obtain high confidence labels from a few labeled instances. Results across eight random forest and neural net models show that the pseudo-labels derived from our SSL method can significantly increase prediction accuracy for 82 CNS tumors and 9 normal controls.ConclusionsThe proposed combination of semi-supervised technique and multinomial logistic regression holds the potential to leverage the abundant publicly available unlabeled methylation data effectively. Such an approach is highly beneficial in providing additional training examples, especially for scarce tumor types, to boost the prediction accuracy of supervised models.

Dataset Information

DNA methylation-based classification of central nervous system tumours.

Publications

DNA methylation-based classification of central nervous system tumours.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets