DNA methylation signatures predict cytogenetic subtype and outcome in pediatric acute myeloid leukemia (AML)

Krali, O, Palle, J et al. 2021

In this GitHub repository, some of data, the Jupyter notebook and python scripts used to produce the paper's results are available and ready to use. For research purposes only that align to investigation of epigenetic signatures implicated on AML. Raw idat files are available upon request (DOI: 10.17044/scilifelab.14666127)(unavailable until the manuscript is published).

Workflow

1. Unzip the raw data given after request
2. Process the data with the Methylprep Python Library
3. Data Preparation
4. Data Visualization
5. Feature Selection
6. Model Selection & Optimization for Subtype Prediction
7. Fit the model on all training data set
8. Predict subtype on the test set
9. Fit the model on all samples of known subtype
10. Permutation analysis
11. Important CpG signatures per Subtype
12. Survival Analysis