Sarcoma DNA-Methylation Classifier

Overview

The sarcoma DNA-methylation classifier (Koelsche et al.) is a machine-learning classifier trained on genome-wide DNA methylation profiles (Infinium EPIC or 450k arrays) from a large reference cohort of sarcoma subtypes. It assigns a tumor’s methylation profile to one of several defined sarcoma methylation classes and returns a calibrated prediction score. A high score (typically >0.9) indicates reliable subclass assignment; low scores or “no class match” indicate an epigenetically novel or unclassifiable entity.

Used by

  • Applied to a urinary bladder EWSR1::BEND2 fusion sarcoma using Infinium MethylationEPIC v2.0 data (Koelsche et al. v13 classifier); the tumor did not match any sarcoma classifier subclass, indicating a novel or uncharacterized epigenetic entity PMID:28199314

Notes

  • Reference classifier: Koelsche et al., published by the German Cancer Research Center (DKFZ) group.
  • Failure to classify (no class match) may indicate a genuinely novel entity, sample quality issues, or admixture with normal tissue.
  • Often interpreted alongside the brain-tumor methylation classifier for tumors with ambiguous lineage.

Sources

This page was processed by crosslinker on 2026-05-14.