Using Multiple Drug Similarity Networks to Promote Adverse Drug Event Detection

Instruction

• Download the Signal Scores and Drug Similarity Matrices from here.
• Copy the downloaed Signal_Scores and Similarity_Matrices directory into the cloned repository.

Running example

python run.py --input Signal_Scores/FAERS_signal_data_BCPNN --method BCPNN --year 22 --quarter 2 --eval_metrics all --similarity go_mf --output output/results_BCPNN/go_mf/results_22Q2.csv --soc_output output/soc_results_BCPNN/go_mf/soc_results_22Q2.csv --split True

Parameters

• --input, input original signal scores files.
• --method, signal detection algorithm (i.e., MGPS, BCPNN).
• --year, year of the signal data file in 'YY' format.
• --quarter, quarter of signal data file (i.e., 1, 2, 3, 4).
• --eval_metrics, evaluation metrics (i.e., AUC, AUPR, Precision, Recall, etc.) choices=['all', 'specificity-sensitivity'].
• --similarity, which drug similarity to use choices=['chem', 'atc', 'sw', 'go_bp', 'go_cc', 'go_mf'].
• --output, output file path for the result.
• --soc_output, output file path for the result of MedDRA SOC.
• --split, whether to split entire dataset into validation set and testing set (i.e., True/False).