The repository contains code for the following tasks included in A prognostic neural epigenetic signature in high-grade glioma [1]:
| Task | Description | Source |
|---|---|---|
| An example run of the signature classifying the neural groups | Contains code to run the fitted logistic regression model on an example IDAT | run neural classification |
| DNA methylation deconvolution | Code to process and perform DNA methylation deconvolution. Please cite Moss et al. (2018) [2] if you use it. | DNAm_deconv |
| CNV | Wrapper to perform copy number variation analysis using Conumee package across multiple groups | CNV |
| Differential methylation probes | Differential methylation probes and gene set enrichment between the neural groups | DNAm_DMP |
| Optimal number of clusters | Overcluster the clinical cohort to find if cluster size > 2 is significantly separable with respect to overall survival | neural_group_over_cluster |
| Signature classifying the neural groups | Contains code to stratify neural groups based on DMP probes between low and neural groups. Include trained logistic regression model | neural_group_classification |
| Mutation analysis | Code to generate oncoplot on mutations | Oncolplot.ipynb |
| RNA groups | Compare correspondence between RNA GBM subgroups and neural subgroups for paired DNAm-RNA TCGA data | TCGA |
| WGCNA | Code to perform WGCNA analysis on paired proteomics data, also includes geneset and cell type enrichment | WGCNA_proteomics |
[1] Drexler, R., Khatri, R., Sauvigny, T. et al. A prognostic neural epigenetic signature in high-grade glioma. Nat Med (2024). https://doi.org/10.1038/s41591-024-02969-w
[2] Moss, J., Magenheim, J., Neiman, D. et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free DNA in health and disease. Nat Commun 9, 5068 (2018). https://doi.org/10.1038/s41467-018-07466-6