problem with messing up VERSION and ACCESSION

[lingyi-owl]

• pharokka version: 1.7.3
• Python version: Python 3.10.14
• Operating System: Linux

Description

I used pharokka to annotate viral contigs. The run was finished successfully. However, some of the VERSION ids are messed up in the output file while the LOCUS and DEFINITION keep the same with the sequence headers of the input fasta file.

For example, in the below output file, the LOCUS is the same with the input sequence header, but the ACCESSION and the VERSION were changed. I subsequently used phold to further annotate the pharokka output genbank file. Phold would use the VERSION id as the sequence ID. Thus, my phold output has different information with the original input sequence headers. It brings me troubles with tracing information along the analysis. Could you please help me with this problem so that I can run pharokka in the future with large amount of data without correcting the sequence headers later?

Best,
Lingyi

LOCUS       BU_D1_MACV_NODE_1129_length_20832_cov_2.052837 20832 bp    DNA     linear   PHG 21-AUG-2024
DEFINITION  BU_D1_MACV_NODE_1129_length_20832_cov_2.052837.
ACCESSION   BU_D1_MACV_NODE_1129_length_20832_cov_2
VERSION     BU_D1_MACV_NODE_1129_length_20832_cov_2.52837

### What I Did

I ran the following code and it finished successfully.

pharokka.py \
        -i ten_seqs.fasta \
        -o pharokka_output \
        -d pharokka_database \
        -t 94

there is a error message in the begnining of the run:

DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
  __import__('pkg_resources').require('Pharokka==1.7.3')
/miniconda3/envs/pharokka_env/lib/python3.10/site-packages/numpy/_core/getlimits.py:555: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
  setattr(self, word, getattr(machar, word).flat[0])
/miniconda3/envs/pharokka_env/lib/python3.10/site-packages/numpy/_core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
  return self._float_to_str(self.smallest_subnormal)
/miniconda3/envs/pharokka_env/lib/python3.10/site-packages/numpy/_core/getlimits.py:555: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
  setattr(self, word, getattr(machar, word).flat[0])
/miniconda3/envs/pharokka_env/lib/python3.10/site-packages/numpy/_core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
  return self._float_to_str(self.smallest_subnormal)

[gbouras13]

Hi @lingyi-owl ,

The behaviour of the gff-gbk conversion program inside pharokka (bcbio-gff) will try and convert the pharokka gff to a genbank based on the genbank format.

https://www.ncbi.nlm.nih.gov/genbank/samplerecord/

I think there is no issue with the ACCESSION, as genbank accessions will omit everything before the first '.' .

GenBank records usually are like e.g "{ACCESSION}.{VERSION_NUMBER}"

Which is the behaviour for your contig.

VERSION is strange, it is cutting out the 0 - my guess is that is parses '052837' as '52837' (and e.g. would parse 01 as 1). I will have a look and may make an issue in bcbio-gff but I'm inclined to not make any changes regardless.

To be honest, in general, I'd advocate renaming your contigs before running pharokka from the complicated/long spades headers to something easier for software to parse.

In my opinion, the problem is really when we run something like pharokka, we are trying to do a bit too much with the GenBank format (which isn't optimal!). So that should mitigate the problem.

George

[estebano11]

Dear George,
I am having the same message, I successfully run pharokka but every time have the same message even if I just run it with -V:
pharokka.py -V
I got this message:
/home/bioinfo/.local/lib/python3.10/site-packages/numpy/_core/getlimits.py:555: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero. setattr(self, word, getattr(machar, word).flat[0]) /home/bioinfo/.local/lib/python3.10/site-packages/numpy/_core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero. return self._float_to_str(self.smallest_subnormal) /home/bioinfo/.local/lib/python3.10/site-packages/numpy/_core/getlimits.py:555: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero. setattr(self, word, getattr(machar, word).flat[0]) /home/bioinfo/.local/lib/python3.10/site-packages/numpy/_core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero. return self._float_to_str(self.smallest_subnormal) 1.7.3
python version = 3.10.14
numpy = 2.1.1

Is something wrong in my configuration?
Thank you

Esteban

[gbouras13]

Hi @estebano11 - this is no bug in pharokka, just a warning, so it is not a big deal and I would just ignore it. There is nothing wrong with your configuration.

I am pretty sure it is caused by numpy v2. You can remove this warning by installing numpy v1 e.g. pip install numpy==1.26.4 I would assume.

George

[estebano11]

I am pretty sure it is caused by numpy v2. You can remove this warning by installing numpy v1 e.g. pip install numpy==1.26.4 I would assume.

George

Dear George,
Thank you, it solved the warming message.

Esteban