Quantification throwing error - missing reference? about arcashla-quant HOT 6 CLOSED

Hello! Thanks for your interest in our tool.

@Michael-Geuenich you'll want to run the "customize" script before the "quant" script, which will construct the missing reference. Use the genotype.json from the "genotype" step:

arcasHLA customize --transcriptome chr6 --genotype sample.genotype.json -o ref

The README will be updated to make this clearer.

@davetang you're correct, this error is caused by a depreciated object in the latest Biopython release. For a quick fix, you can revert to an older version of Biopython (1.77 or earlier) until we can provide a workaround in an upcoming release.

from arcashla-quant.

I'm also interested in the quant functionality and I have the same problem as @Michael-Geuenich. The extract and genotype steps worked fine.

Also just for your information, when I try to run the customize step, I get the following error:

Traceback (most recent call last):
  File "/home/dtang/src/test/arcasHLA-quant/scripts/customize.py", line 43, in <module>
    from Bio.Alphabet import generic_dna
  File "/home/ngs/Analysis/tools/arcashla_venv/lib/python3.9/site-packages/Bio/Alphabet/__init__.py", line 20, in <module>
    raise ImportError(
ImportError: Bio.Alphabet has been removed from Biopython. In many cases, the alphabet can simply be ignored and removed from scripts. In a few cases, you may need to specify the ``molecule_type`` as an annotation on a SeqRecord for your script to work correctly. Please see https://biopython.org/wiki/Alphabet for more information.

I guess I need to use an older version of Biopython.

from arcashla-quant.

Hi @tpereachamblee,

thank you for the comment. I downgraded my Biopython version and could run the customize subcommand. However, I'm getting the following error after running:

arcasHLA quant --ref ref/my_sample -o test my_sample.extracted.1.fq.gz my_sample.extracted.2.fq.gz

Traceback (most recent call last):
  File "/home/dtang/Analysis/tools/arcasHLA-quant_down/scripts/quant.py", line 300, in <module>
    baf = allele_results[gene]['allele1_count'] / (allele_results[gene]['allele1_count'] + allele_results[gene]['allele2_count'])
ZeroDivisionError: division by zero

May you let me know what the problem could be?

Thanks again,
Dave

from arcashla-quant.

Hi @tpereachamblee,

thanks for the additional clarification. I've run the command exactly as you specified within the arcasHLA-quant docker container. Unfortunately I am still getting the following error:

Traceback (most recent call last):
  File "/home/arcasHLA-quant/scripts/customize.py", line 299, in <module>
    build_custom_reference(subject, genotype, args.grouping, args.transcriptome, temp)
  File "/home/arcasHLA-quant/scripts/customize.py", line 94, in build_custom_reference
    with open('dat/ref/allele_groups.p','rb') as file:
FileNotFoundError: [Errno 2] No such file or directory: 'dat/ref/allele_groups.p'

Any insights on what the issue could be?

from arcashla-quant.

Hi @Michael-Geuenich,

I had the same error because I wasn't running arcasHLA from where I cloned and prepare the repo. I copied the dat directory from the cloned location into where I was executing arcasHLA and that circumvented that error. But then I got the divide by zero error, so maybe it's not advisable to copy the folder over. That said, let me know how it works for you!

Cheers,
Dave

from arcashla-quant.

Thank you both for your helpful feedback. I believe @davetang's solution should work for you @Michael-Geuenich the customize script needs to be updated work from outside of the arcasHLA folder (presently it does not).

@davetang as far as your issue running quant, I am still looking into it and will comment here once I think I know what is going on.

from arcashla-quant.