This repository contains the code for
- Make predictions for new patients.
- Generating analysis and plots.
-
data
: folder containing the data used for the analysis (empty as data cannot be shared) -
figures
: folder containing figures used for the manuscript -
model_development
: R code and model fitting objects (several models are included based on the different models fitted for the article) -
model_predictions
: Vectors containing MODY probability estimations for all of the datasets used (_full.rds
represents the full posterior dataset, its omission represents the summary results (2.5%, mean, 97.5%)) -
new_data_predictions
: R code + data + Excel document for making predictions in new datasets -
bootstrap_riley_simulation
: R code for a simulation study on stability of model predictions based on bootstrapped datasets.
discrimination_curves.R
: this file contains the R code for calculating and plotting ROC and Precision-recall curves.bootstrap_validation_analysis.R
: this file contains the R code for making calibration plots of bootstrapped datasets.model_convergence.R
: this file contains the R code for checking model convergence in the main models used for this analysis.calibration_plots.R
: this file contains the R code for creating the calibration plots for the datasets used.CPRD_analysis.R
: this file contains the R code for plotting density plots for probability of MODY.antibody_sensitivity_analysis.R
: this contains the R code for plotting probabilities of MODY based on different antibody information availability.