PrankWeb is web interface based on P2Rank standalone method. For batch-processing, it is recommended to download standalone version (link) of P2Rank and run experiments locally.
PrankWeb classification model has been trained and validated on publicly available datasets. They are available at https://github.com/rdk/p2rank-datasets. The conservation pipeline uses HSSP database as well as SwissProt and TrEMBL databases.
PrankWeb wouldn't exist if not for these projects.