###################################################################################################
Some code and resources of the CASAVA paper.Resources are listed below.
For any problem concerning the data and the code, please contact [email protected].

1) Official release of CASAVA score:
  https://github.com/zhanglabtools/CASAVA/releases.

2) CASAVA web application: 
  http://zhanglabtools.org/CASAVA.
  
3) Training-testing sets, trained models, predictions and raw evaluation results can be downloaded from Zenodo: 
  https://zenodo.org/record/4365899#.X-cB6tgzaUk
  
4) Training, testing sets, trained models, predictions can be also downloaded from Baidu Cloud: 
  https://pan.baidu.com/s/1loDDBiLg6youoTkfS_AlCg 
  passwork：on5p 

5) The main scripts used in our experiments with detailed doucuments.
  https://github.com/zhanglabtools/CASAVA

  CASAVA_library.R contains the commonly used functions.
  train_CASAVA_script.R is used to train CASAVA models.
  train_TrCASAVA_script.R is used to train disease-specific and TrCASAVA models.
  train_Zeroshot_script.R is used to train CASAVA models in the scenario of no disease-specific training data available.
 
 For some other scripts in development, please do not use them unless necessary.