This repository contains jupyter notebooks used in "The frequency of somatic mutations in cancer predicts the phenotypic relevance of germline mutations"

model evaluation contains notebooks used for nested cross-validation of our candidate models.
feature importance contains notebooks used for evaluating the importance of the features in our dataset for each model.
multiphenotype contains notebooks used to fit the final, multiphenotype model which is used in the DISCAN app. The results from this notebook are not included in the repository because they exceed github's maximum file size.

Click here to visit the repository containing the shiny app which implements the multiphenotype model.