eval_imputation_performance {Melissa} | R Documentation |
eval_imputation_performance
is a wrapper function for
computing imputation/clustering performance in terms of different metrics,
such as AUC and precision recall curves.
eval_imputation_performance(obj, imputation_obj)
obj |
Output of Melissa inference object. |
imputation_obj |
List containing two vectors of equal length, corresponding to true methylation states and predicted/imputed methylation states. |
The 'melissa' object, with an additional slot named 'imputation', containing the AUC, F-measure, True Positive Rate (TPR) and False Positive Rate (FPR), and Precision Recall (PR) curves.
C.A.Kapourani C.A.Kapourani@ed.ac.uk
create_melissa_data_obj
, melissa
,
impute_test_met
, filter_regions
,
eval_imputation_performance
,
eval_cluster_performance
# First take a subset of cells to efficiency # Extract synthetic data dt <- melissa_synth_dt # Partition to train and test set dt <- partition_dataset(dt) # Create basis object from BPRMeth package basis_obj <- BPRMeth::create_rbf_object(M = 3) # Run Melissa melissa_obj <- melissa(X = dt$met, K = 2, basis = basis_obj, vb_max_iter = 10, vb_init_nstart = 1, is_parallel = FALSE, is_verbose = FALSE) imputation_obj <- impute_test_met(obj = melissa_obj, test = dt$met_test) melissa_obj <- eval_imputation_performance(obj = melissa_obj, imputation_obj = imputation_obj) cat("AUC: ", melissa_obj$imputation$auc)