Correlogy v0.1a

Gene Co-Occurrence Across Phylogeny See developer doc: https://docs.google.com/document/d/1jad90RfWmCfFbAY6HheRkbyQ1qc92bQbZlQKdDB_9KE/edit?usp=sharing

How to run

Choose BASH or IDE (eg Spyder)

1. BASH: Must set 'use_IDE' variable to 'False' in main.py

   • minimal example: python3 main.py -i input_gene_list
   • full example: some text
   • usage: correlog [-h] -i INPUT [-o OUTPUT] [--evalue EVALUE] [--entrez ENTREZ] [-s SKIPBLAST]
   • get help, list all flag defaults: python3 main.py -h

2. IDE: Must set 'use_IDE' variable to 'True' in main.py

   • Hardcode flag settings in housekeeping.py --> parse_input() --> arg_dict dictionary

Known bugs

1. NCBI BLASTing can sometimes time out, resulting in an empty XML file in 01_BLAST_results. Workaround is to use '-s' flag to skip BLAST query and put your own BLAST results in XML format (one file per query) into 01_BLAST-results.

2. blast_parse.py --> create_pa() does 'flattening' of pa_df via 'pa_df_rows = pd.unique(merged_df[pa_df_columns].values.ravel()).tolist()'. This returns some hits as "NaN", not sure why probably related to flattening.

To do

in housekeeping.py --> create_folders()

1. validate input file exists (otherwise throw error)
2. validate output directory exists (otherwise throw error)