Related Experiment Video
Updated: Jul 17, 2025
06:53
Detection and Monitoring of Tumor Associated Circulating DNA in Patient Biofluids
Published on: June 8, 2019
8.7K
EpiGe: A machine-learning strategy for rapid classification of medulloblastoma using PCR-based methyl-genotyping
Soledad Gómez-González1,2, Joshua Llano2,3,4, Marta Garcia1,2
1Laboratory of Developmental Tumor Biology, Institut de Recerca Sant Joan de Déu, Pediatric Cancer Center Barcelona, Hospital Sant Joan de Déu, Barcelona, Spain.
Iscience
|September 4, 2023
Summary
A new machine-learning system classifies medulloblastoma molecular groups using qPCR methylation data. This user-friendly web-app, EpiGe-App, offers rapid, accessible brain tumor classification for improved treatment strategies.
Area of Science:
- Oncology
- Genetics
- Bioinformatics
Background:
- Medulloblastoma molecular classification is crucial for effective treatment.
- Array-based DNA methylation profiling is a powerful but often unavailable tool.
- There is a need for accessible methods for medulloblastoma classification.
Purpose of the Study:
- To develop a machine-learning decision support system (DSS) for classifying medulloblastoma molecular groups.
- To enable classification directly from quantitative PCR (qPCR) data.
- To create a user-friendly web-application for automated interpretation and prediction.
Main Methods:
- Utilized a previously validated six-cytosine signature with subgroup-specific methylation profiles.
- Developed a methyl-genotyping assay for determining cytosine methylation status via qPCR.
- Integrated the DSS into a web-application named EpiGe-App for automated analysis.
Main Results:
- Successfully classified principal molecular groups (WNT, SHH, non-WNT/non-SHH) from qPCR data.
- Demonstrated the feasibility of using a reduced set of markers for accurate classification.
- Developed a web-application for accessible and automated interpretation of methylation data.
Conclusions:
- The developed DSS and EpiGe-App provide a comprehensive and rapid approach for medulloblastoma classification.
- This method leverages readily accessible qPCR equipment and a user-friendly interface.
- Facilitates timely and accurate molecular subtyping for improved clinical decision-making in medulloblastoma treatment.
