aka Alternating Imputation and Correction Method
2018 Zhiyue Tom Hu, Yuting Ye, Patrick A. Newbury, Haiyan Huang and Bin Chen
Full paper available at: https://www.biorxiv.org/content/early/2018/08/07/386896
If you are simply interested in fetching the corrected and/or the raw data, please see our corrected data and raw data in /data.
Current version is 0.1. We will be upgrading soon, stay tuned!
Please make sure you have all dependencies installed in your python environment.
You can run it in the following way:
python aicm.py /data/original_data/GDSC_CTRP_auc.csv /data/original_data/CTRP_GDSC_auc.csv /imputed [Options]
The first two arguments are the path to your desired-to-correct .csv files. The third one is where the output will be automatically saved to. They will be named as df1_imp.csv and df2_imp.csv.
If you would like to tune parameters, please use python aicm.py /data/original_data/GDSC_CTRP_auc.csv /data/original_data/CTRP_GDSC_auc.csv /imputed -h to find out what option is, what do they mean and their default values.
An example would be:
python aicm.py /data/original_data/GDSC_CTRP_auc.csv /data/original_data/CTRP_GDSC_auc.csv /imputed --dropping_rate 0.1 to change the dropping rate to 0.1 instead of default 0.05.