Sequencing of cell-free DNA in the blood of cancer patients (liquid biopsy) provides attractive opportunities for early diagnosis, assessment of treatment response, and minimally invasive disease monitoring. To unlock liquid biopsy analysis for pediatric tumors with few genetic aberrations, we introduce an integrated genetic/epigenetic analysis method and demonstrate its utility on 241 deep whole genome sequencing profiles of 95 patients with Ewing sarcoma and 31 patients with other pediatric sarcomas. Our method achieves sensitive detection and classification of circulating tumor DNA in peripheral blood independent of any genetic alterations. Moreover, we benchmark different metrics for cell-free DNA fragmentation analysis, and we introduce the LIQUORICE algorithm for detecting circulating tumor DNA based on cancer-specific chromatin signatures. Finally, we combine several fragmentation-based metrics into an integrated machine learning classifier for liquid biopsy analysis that is tailored to cancers with low mutation rates while exploiting widespread epigenetic deregulation. Clinical associations highlight the potential value of cfDNA fragmentation patterns as prognostic biomarkers in Ewing sarcoma. In summary, our study provides a comprehensive analysis of circulating tumor DNA beyond recurrent somatic mutations, and it renders the benefits of liquid biopsy more readily accessible for childhood cancers.

You can find the associated publication at Nature Communications:

NEWS: Please note that we have recently released a new version of our tool LIQUORICE. See here for more information.

Study overview

A comprehensive combination of fragment-based analyses allows temporally resolved tumor monitoring and sensitive detection and classification of tumors based on their epigenetic signature


This section provides access to pre-processed data and analysis results underlying the presented analysis of Ewing sarcoma cfDNA. The raw sequencing data have been deposited at the European Genome-phenome Archive (EGA) under accession number EGAS00001005127 - this data is available under a controlled access regimen to ensure the protection of personally identifiable data.

Name Description and link
ichorCNA's output files for all samples (CNA's per called segment and per 500kb bin) .tar.gz file, download here
Fragment size distribution for all samples (Picard CollectInsertSizeMetrics) .tar.gz file, download here
Regional fragment size: S/L ratio in 100kb bins log2(S/L) ratio per bin per sample, excluding CNA affected bins, download here
z-scored log2(S/L) ratio vs. controls, including CNA affected bins, download here
raw, GC-corrected nr. of long fragments (151-220 bp) per bin, per sample, download here
raw, GC-corrected nr. of short fragments (100-150 bp) per bin, per sample, download here
Fragment coverage at regions-of-interest .bigwig files for all samples of this study, download here
Machine learning based detection of EwS and distinction from other sarcomas Collected input feature-sets for ML , download here

We have recently released an improved, updated version of LIQUORICE! You can read the documention, including installation and usage instructions here. The source code is hosted here. We still provide the version of LIQUORICE we originally used for this pulication below (LIQUORICE source code; Usage example; see below for installation instructions). However, we would recommend you to use our more user-friendly updated version if you want to use LIQUORICE on your own data.


This section provides access to the source code underlying the presented analysis of Ewing sarcoma cfDNA. For convenience, we provide .html containing the source code below. These files can easily be viewed in the browser. If you want to re-run the analysis on your own system, please download the complete code directory here. Then, if not already installed on your system, please install conda, create a new environment and install the necessary dependencies:
conda env create -f environment.yml
conda activate ews_cfdna (or "source activate ews_cfdna" for older conda versions)
Now, you should be able to execute the code (provided as .html and .ipynb files) in the individual folders, corresponding to the figures in the paper. Please contact in case you experience any issues.

Analysis & link
Source code for figure 2
Source code for figure 4
Source code for figure 5
Source code for figure 6


If you use these data in your research, please cite:

Peter Peneder*, Adrian M. Stütz*, Didier Surdez, Manuela Krumbholz, Sabine Semper, Mathieu Chicard, Nathan C. Sheffield, Gaelle Pierron, Eve Lapouble, Marcus Tötzl, Bekir Ergüner, Daniele Barreca, André F. Rendeiro, Abbas Agaimy, Heidrun Boztug, Gernot Engstler, Michael Dworzak, Marie Bernkopf, Sabine Taschner-Mandl, Inge M. Ambros, Ola Myklebost, Perrine Marec-Bérard, Susan Ann Burchill, Bernadette Brennan, Sandra J. Strauss, Jeremy Whelan, Gudrun Schleiermacher, Christiane Schaefer, Uta Dirksen, Caroline Hutter, Kjetil Boye, Peter F. Ambros, Olivier Delattre, Markus Metzler, Christoph Bock§ & Eleni M. Tomazou§ (2021). Multimodal analysis of cell-free DNA whole genome sequencing for pediatric cancers with low mutational burden. Nature Communications, DOI: 10.1038/s41467-021-23445-w.

* These authors contributed equally to this work
§ Co-last author / These authors jointly directed this work