IndexError: list index out of range #38
Description
Title: IndexError: list index out of range during funannotate predict
Description:
I am trying to run Funannotate v25.11.1 with Python 3.12 for gene prediction using previously generated training files. During the execution of funannotate predict, I encounter the following error:
Traceback (most recent call last):
File "funannotate2", line 6, in
sys.exit(main())
File "site-packages/funannotate2/main.py", line 26, in main
predict(args)
File "site-packages/funannotate2/predict.py", line 305, in predict
source = list(sources)[0].lower()
IndexError: list index out of range
Command used (paths anonymized):
funannotate predict
--input-dir /path/to/predict
--fasta /path/to/clean_assembly.fasta.masked
--species "Chlorella sorokiniana"
--params /path/to/train_results/Chlorella_sorokiniana.params.json
--out predict_output
Relevant log excerpt (anonymized):
Training parameters loaded for Chlorella_sorokiniana (augustus, glimmerhmm, snap, genemark)
Existing transcript alignments found, will reuse and continue
Parsed ~496k proteins from reference sources
Notice: the log shows a suspicious concatenation in protein source paths, e.g.
/path/to/reference/Chlorella_sorokiniana_protein.faa/n/path/to/uniprot_sprot.fasta
It looks like an extra /n/ is inserted between file paths.
Additional info:
All paths have been verified to exist
Training (funannotate train) completed successfully
The error occurs after attempting to reuse existing protein alignments
Question:
Has anyone encountered this IndexError: list index out of range during predict? Could this be caused by the erroneous /n/ inserted between protein files? Any guidance on fixing this would be greatly appreciated.