Code from: "Computational measurable residual disease assessment in acute myeloid leukemia using mixture models"
This repository uses both R (flow cytometry pre-processing) and Python (modeling) together. The code is structured as follows:
| Script | Function |
|---|---|
| 01-preprocess.Rmd | Pre-processing of flow cytometry data (BLAST110, LAIP29 and RBM18 datasets) |
| 02-GMMclf.ipynb | Blast prediction models + benchmark |
| 03-WBC.ipynb | WBC prediction models |
| 04a-refGMM.ipynb | Reference GMM model training |
| 04b-refGMM-annotation.ipynb | Generation of model component CSV file for FCS annotation |
| 04c-refGMM-labeling.Rmd | Construction of annotated FCS files |
| 05-benchmark.ipynb | Benchmark of cMRD methods |
| 06a-timing-R.Rmd | Timing of R scripts |
| 06b-timing-Python.ipynb | Timing of Python scripts |
Datasets are available from Zenodo: https://zenodo.org/records/11046402