Sarcoma methylation classifier
Bone and soft-tissue tumour classification, from arrays or sequencing.
How it works
Methylation-based classification of sarcomas
Previously distributed under the MolecularNeuropathology banner, our sarcoma methylation classifier now runs on our own platform and as a licensed on-premise installation.
A well-annotated reference cohort
The classifier is trained on 4,377 methylation profiles covering 120 subclasses of bone and soft-tissue tumour. A machine-learning model compares each new case against that cohort and returns a score reflecting its resemblance to each class.
A calibrated confidence score
Every class receives a score between 0 and 1, so borderline cases are visible as borderline rather than presented as certainty. Sarcoma subtyping is exactly where that matters: many entities are morphologically overlapping and hard to separate on histology alone.
Robust on difficult material
The cancer methylome combines somatically acquired alterations with features indicating the cell of origin. Methylation profiling is highly reproducible and often succeeds on small or degraded samples where other approaches do not.
Two-tier hierarchical output
Results are returned as methylation subclass and class, alongside a copy number profile and quality-control metrics.
Specifications
Version 13.1
Coverage
- 120 subclasses
- 93 classes
- Reference cohort of 4,377 methylation profiles
Performance
- Two-tier hierarchical output
- Proposed evidence level annotation
- Confidence score for every call
In use
- Over 200,000 cases analysed to date
- More than 500 institutions worldwide
- On-premise version runs on standard laptops
- Installation guide included
Workflow
From sample to report
- Sample collection: fresh frozen or FFPE, tumour content at least 70%
- DNA extraction
- Raw data generation
- Classification
- PDF and HTML report
Every report includes
- Quality control metrics
- Tumour classification with confidence score
- Copy number variation analysis
Supported platforms
Platform-agnostic: now compatible with sequencing data
The classifier is not tied to one instrument. One consistent model handles both array and sequencing input, so a laboratory can add or change technology without changing classifier and without re-validating against a separate model.
Methylation arrays
- Illumina Infinium MethylationEPIC v2
- Illumina Infinium MethylationEPIC v1
- Illumina HumanMethylation450 (450k)
Upload unprocessed IDAT files and the sample is classified automatically.
Sequencing
- Nanopore whole-genome sequencing
- Nanopore adaptive sampling
- Whole-genome bisulfite sequencing (WGBS)
- EM-seq
- TAPS+
- Illumina 5-base
- biomodal 6-base
- Targeted methylation panels
Sequencing-derived methylation data is classified with the same model, with no ad-hoc per-platform variant.
Key publications
The science behind the classifier
Nature Communications · 2021
Koelsche et al. Sarcoma classification by DNA methylation profiling
Established that methylation profiling can resolve bone and soft-tissue tumours into robust, reproducible classes, the foundation this classifier is built on.
medRxiv · 2025
Jäger et al. Advancing sarcoma diagnostics with expanded DNA methylation-based classification
Describes the expanded reference cohort and class structure underlying version 13.1.
Access and licensing
Two ways to use the sarcoma classifier
The classifier is available for research use on our RUO platform at app.epignostix.com, free for academic research and scientific use. Register there and log in once your account is approved.
For routine diagnostic use, the on-premise version is licensed for local installation: fully offline, with installation and validation support and complete documentation. Contact licensing@epignostix.com.
