Random Forest algorithm pipeline for classification of PanNETs.
In this repository, we tried to reproduce the work done by Capper et al. (http://www.nature.com/doifinder/10.1038/nature26000) but for classification of Panceratic Neuro Endocrine Tumors (PanNETs). We mainly based our scripts from https://github.com/mwsill/mnp_training and adapted them to our data.
The scripts are not reproducible for now. Ideally they should be adapated in order to create a pipeline for training and testing other data. In the future, it
could be great to also reproduce this work in Python with the sckit-learn package. For now everything is in R.
File 043_panet_classification_summary.Rmd should be ignored. It used to be the main script but we change it to multiple small R scripts. It has not been updated since.
Results can be seen from the two html files: 043_panet_classification_5_classes_summary.html and 043_panet_classification_5_classes_summary.html.
043_panet_classification_5_classes_summary.html should be read first.