Error in makedatabase_struct.py about metatag

[Robaina]

Hi,

It seems like a new version of the library_pyfastx_ and a old index of the fasta file (created by pyfastx) could be the cause (see the post here: lmdu/pyfastx#9)

Please, check if you have an old index file for the "input_fasta" in the directory where you are currently working, if so delete it and try again (or start the program in a fresh, empty directory)

Let me know how it goes

[gecko1990]

Hi,

I tried to do so by removing everything in the folder (rm -rf *), but it didn't work. I also deleted my old environment and reinstalled from scratch to check if it was a version problem, but the problem persists

[gecko1990]

Update:

I tried the same procedure with a different gene (Amt, TIGRFAM TIGR00836) to check if it was related to the makedatabase_struct.py or the KEGG format. The error appears again:

(traits) rlaso@elbrus:/data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Amt/data/reference_data_Amt_TIGR00836$ python3 /data/mcm/rlaso/Programs/TRAITS/code/makedatabase.py --in /data/mcm/databases/TRAITs_reference_database/final_ref_database.fasta --outdir /data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Amt/data/reference_data_Amt_TIGR00836/ --hmm /data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Amt/hmms/TIGR00836.1.HMM --relabel --prefix "ref_TIG"
* Making peptide-specific reference database...
Running Hmmer...
Parsing Hmmer output file...
Filtering Fasta...
RuntimeError: get seq count and length error

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/data/mcm/rlaso/Programs/TRAITS/code/makedatabase.py", line 118, in <module>
    main()
  File "/data/mcm/rlaso/Programs/TRAITS/code/makedatabase.py", line 81, in main
    filterFASTAByHMM(
  File "/data/mcm/rlaso/Programs/TRAITS/code/phyloplacement/database/manipulation.py", line 118, in filterFASTAByHMM
    filterFASTAbyIDs(input_fasta, record_ids=hmmer_hits.id.values,
  File "/data/mcm/rlaso/Programs/TRAITS/code/phyloplacement/database/manipulation.py", line 76, in filterFASTAbyIDs
    fa = pyfastx.Fasta(input_fasta)
SystemError: <class 'Fasta'> returned a result with an error set

I guess that means the error is in the database or on my specific environment

[gecko1990]

In principle the md5sum is the correct one.
5fb8268b59b222a9cd035c46bcf4d031 final_ref_database.fasta

[Robaina]

Hi @gecko1990 (sorry for the late reply),

It seems like a new version of the library_pyfastx_ and a old index of the fasta file (created by pyfastx) could be the cause (see the post here: lmdu/pyfastx#9)

Please, check if you have an old index file for the "input_fasta" in the directory where you are currently working, if so delete it and try again (or start the program in a fresh, empty directory)

Let me know how it goes

[gecko1990]

I tried that as referred here (#53 (comment)), but I am unsure what you mean with the working directory.

(traits) rlaso@elbrus:/data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Nxr_Narg/data$ python3 /data/mcm/rlaso/Programs/TRAITS/code/makedatabase.py --in /data/mcm/databases/TRAITs_reference_database/final_ref_database.fasta --outdir /data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Nxr_Narg/data/ --hmm /data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Nxr_Narg/hmms/TIGR01580.1.HMM --relabel --prefix "ref_TIG"        * Making peptide-specific reference database...
Running Hmmer...
Parsing Hmmer output file...
Filtering Fasta...
RuntimeError: get seq count and length error

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/data/mcm/rlaso/Programs/TRAITS/code/makedatabase.py", line 118, in <module>
    main()
  File "/data/mcm/rlaso/Programs/TRAITS/code/makedatabase.py", line 81, in main
    filterFASTAByHMM(
  File "/data/mcm/rlaso/Programs/TRAITS/code/phyloplacement/database/manipulation.py", line 118, in filterFASTAByHMM
    filterFASTAbyIDs(input_fasta, record_ids=hmmer_hits.id.values,
  File "/data/mcm/rlaso/Programs/TRAITS/code/phyloplacement/database/manipulation.py", line 76, in filterFASTAbyIDs
    fa = pyfastx.Fasta(input_fasta)
SystemError: <class 'Fasta'> returned a result with an error set
(traits) rlaso@elbrus:/data/mcm/rlaso/Traits/Phylogenetic_trees/Nitrogen_cycle/Nxr_Narg/data$ ls
hmmer_output.txt

As I show here, once the error occurs, the only thing appearing in the working directory is the hmmer output

[Robaina]

Hi, yes, sorry, I meant the directory where the input fasta is located, in your case:

/data/mcm/databases/TRAITs_reference_database/final_ref_database.fasta

Could you please check if there is any file ending in ".fxi", probably "final_ref_database.data.fxi"?

Thanks

[gecko1990]

Nothing at all. =(

(traits) rlaso@elbrus:/data/mcm/rlaso/Programs/TRAITS$ ls /data/mcm/databases/TRAITs_reference_database/
final_ref_database.fasta
(traits) rlaso@elbrus:/data/mcm/rlaso/Programs/TRAITS$ cd /data/mcm/databases/TRAITs_reference_database/
(traits) rlaso@elbrus:/data/mcm/databases/TRAITs_reference_database$ ls -a
.  ..  final_ref_database.fasta

[gecko1990]

Problem resolved. I didn't have writing permissions for the folder and the index could not be created

[Robaina]

Alright, closing this issue then!