Pan-cancer Graph-based Cancer Detection Using the Cell-free DNA Methylome
Zhao, L.; Zeng, Y.; Abelman, D. D.; Lin, W.; Luo, P.
Show abstract
Motivation: Cell-free DNA methylation provides a minimally invasive signal for early cancer detection and tissue-of-origin prediction. Most methods represent methylation measurements as independent fixed-window features and therefore do not explicitly model relationships among genomic regions. Results: We developed PANGEM (Pan-cancer Graph-based Cancer Detection Using the Cell-free DNA Methylome), a graph-learning framework that represents genomic bins as nodes and integrates CpG context, genomic proximity, and sample-specific methylation similarity in the graph topology. Across five repeated stratified train-test splits, PANGEM achieved the highest mean performance among evaluated methods, with an AUROC/AUPR of 0.997/1.000 for binary cancer detection and macro-AUROC/AUPR of 0.977/0.870 for multiclass tissue-of-origin prediction. In the independent INSPIRE cohort, 72 of 78 cancer cases (92.3%) exceeded the binary classification threshold, and PANGEM correctly classified 9 of 17 head and neck cancer cases (52.9%), the highest accuracy among evaluated methods. Subnetwork analysis further identified recurrent, graph-connected methylation patterns, including a 111-DMR subnetwork with increased methylation in cancer samples.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Rockfish: A Transformer-based Model for Accurate 5-Methylcytosine Prediction from Nanopore Sequencing 93%
- MethylBERT: A Transformer-based model for read-level DNA methylation pattern identification and tumour deconvolution 93%
- Machine learning-based tissue of origin classification for cancer of unknown primary diagnostics using genome-wide mutation features 92%
Similar papers in this journal
- Methylation content sensitive enzyme ddRAD (MCSeEd): a reference-free, whole genome profiling system to address cytosine/ adenine methylation changes 94%
- Leveraging collective regulatory effects of long-range DNA methylations to predict gene expressions and estimate their effects on phenotypes in cancer 93%
- Molecular counting enables accurate and precise quantification of methylated ctDNA for tumor-naive cancer therapy response monitoring 93%
Similar papers in this journal
- Non-invasive multi-cancer detection using DNA hypomethylation of LINE-1 retrotransposons 93%
- cfTrack : Exome-wide mutation analysis of cell-free DNA to simultaneously monitor the full spectrum of cancer treatment outcomes: MRD, recurrence, and evolution 92%
- Clinical Validation of Digital PCR-based ctDNA detection for risk stratification in residual triple negative breast cancer: TRICIA trial results 89%
Similar papers in this journal
- Deep learning inference of cell type-specific gene expression from breast tumor histopathology 93%
- Real-World Benchmarking and Validation of Foundation Model Transformers for Endometrial Cancer Subtyping from Histopathology 92%
- Generalizing AI-driven Assessment of Immunohistochemistry across Immunostains and Cancer Types: A Universal Immunohistochemistry Analyzer 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.