Desktop version of AcRanker AcRanker is a machine learning system developed in python that ranks proteins in a proteome as per their Anti-CRISPR tendencies predicted using sequence features. The method takes as input a proteome in FASTA format and returns a ranked list as per the expected Acr behavior in csv format. Use the following command to generate predictions: python acranker.py <input file name> <output file name> For example, python acranker.py input_proteome.fasta results Following packages are required: Biopython sklearn scipy