Prediction of repeated-dose intravenous ketamine response in major depressive disorder by using the GWAS-based machine learning approach

• step1_split_dataset.py : Randomly divide the initial dataset into six folds.
• step2_feature_selection.py : Calculate random forest importance score based on GWAS result.
• step3_model_construction.py : Model construction.
• plink.sh : Conduct quality control and genome-wide logistic regression in PLINK v.1.9 and encode the the genotype data as 0, 1 or 2.
• models.zip : The models conducted in this study (pickle files).