Dataset Information

Open Chromatin Guided Interpretable Machine Learning Reveals Cancer-Specific Chromatin Features in Cell-free DNA

ABSTRACT: Cell-free DNAs (cfDNAs) are DNA fragments found in blood, originating mainly from immune cells in healthy individuals and from both immune and cancer cells in cancer patients. While cancer-derived cfDNAs carry mutations, they also retain epigenetic features such as DNA methylation and nucleosome positioning. In this study, we examine nucleosome enrichment patterns in cfDNAs from breast and pancreatic cancer patients and find significant enrichment at open chromatin regions. Differential enrichment is observed not only at cancer cell type specific ATAC-seq peaks but also at CD4? T cell specific peaks, suggesting both tumor- and immune-derived contributions to the cfDNA signal. To leverage these patterns, we apply an interpretable machine learning model (XGBoost) trained on cell type specific open chromatin regions. This approach improves cancer detection accuracy and highlights key genomic loci associated with the disease state. Our pipeline provides a robust and interpretable framework for cfDNA-based cancer detection.

ORGANISM(S): Homo sapiens

PROVIDER: GSE279542 | GEO | 2025/09/08

REPOSITORIES: GEO

ACCESS DATA

Dataset's files

Source:

			Action	DRS
		Other

Items per page:

1 - 1 of 1

Similar Datasets

Project description:Background: Gastric cancer (GC) remains the third leading cause of cancer death worldwide due to the absence of sensitive and specific biomarkers for its early detection. 5-hydroxymethylcytosine (5hmC)-enriched gene profiles and regions show tissue-specific and tumor specific, and cell-free DNA (cfDNA) 5hmC modification feature sequencing provides a robust tool for gastric cancer detection.Methods: A matched case‒control study design with 50 gastric cancer patients and 50 controls was performed to sequence the 5hmC modification of cfDNA. Significantly differential 5hmC modification genes were identified to construct a gastric cancer diagnostic model. The data set from Gene Expression Omnibus consisted of 61 gastric cancer patients, 90 healthy individuals and 22 benign gastric disease controls was used as an external testing set to test the robustness of the diagnostic model. Results: The vast majority of 5hmC peaks were distributed in the gene body, accounting for more than 90% of the total 5hmC peaks in both the GC and control groups. The diagnostic model was developed based on five different 5hmC modification genes, FBXL7, PDE3A, TPO, SNTG2 and SXBP5. The model could clearly distinguish gastric cancer patients from controls in the training (AUC=0.95, sensitivity=88.6%, specificity=94.3%), validation (AUC=0.87, sensitivity=73.3%, specificity=93.3%) and testing (AUC=0.90, sensitivity=81.9%, specificity=90.2%) sets. The predicted risk scores of the controls were significantly lower than those of GC patients in our own data (P<0.001) or GEO external testing data (P<0.001), but no significant difference was observed between different TNM stage groups (P=0.09 and 0.66). Furthermore, the risk scores of healthy and benign gastric disease controls in the testing set were also not different (P=0.10). Conclusions: The characteristics of 5hmC in cfDNA are specific to GC patients, and the diagnostic model constructed by the 5hmC features of five genes could effectively identify GC patients.

Dataset Information

Open Chromatin Guided Interpretable Machine Learning Reveals Cancer-Specific Chromatin Features in Cell-free DNA

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets