fusion_collate_info.py script has bugs and its not working.

[unique379r]

I am trying to use fusion script and then sqanti3_qc and finally trying to classify the fusions similar to your tutorial (https://github.com/Magdoll/cDNA_Cupcake/wiki/Best-practice-for-fusion-transcript-finding). But at the end its not working.

Command

fusion_collate_info.py --genome $reffa isoform_fusion isoform_fusion_classification.txt $refgtf

Error

!st it complains about having isoform_fusion.ref.fa which i never got from cupcake but anyway i just converted the rep.fq to rep.fa and then ran it again but again got some errors:

/stornext/snfs5/next-gen/scratch/rupesh/Apps/envs/cupcake/bin/fusion_collate_info.py:171: SyntaxWarning: "is not" with a literal. Did you mean "!="?
  'ReadCountScore': count_d[pbid]*(10**6)/total_fl_count  if count_d[pbid] is not 'NA' else 'NA',

  File "/stornext/snfs5/next-gen/scratch/rupesh/Apps/envs/cupcake/bin/fusion_collate_info.py", line 229, in <module>
    collate_info(args.fusion_prefix, args.class_filename, args.gtf_filename,
  File "/stornext/snfs5/next-gen/scratch/rupesh/Apps/envs/cupcake/bin/fusion_collate_info.py", line 95, in collate_info
    count_d[r['pbid']] = int(r['count_fl'])
KeyError: 'count_fl'

Please help.