middle_initial in submol.yaml

[ncbi]

middle_initial in submol.yaml

[azat-badretdin]

Update:

middle_initial in contact_info leads to a crash

We added middle_initial support in contact_info in 2020-02-06.build4373 release

[TheBigFatTony]

I accidentally published the issue before completing it. Do you have enough information now?

[azat-badretdin]

Yes. Thank you! I am adding internal ticket labels right now.

[azat-badretdin]

Also, it would be great if pgap gave specific error messages

Understood. Sometimes it is easy to fix them, sometimes it is harder. Feel free to open a new issue for particular error messages that need work.

[thibaudnis]

The documentation regarding contact_info has been updated. Thank you for reporting the problem!
PS: we are working on the other issues you mentioned.

[TheBigFatTony]

thanks for the rapid response!

it might be useful to programmatically formulate the requirements in pgap.py and to check them before running the docker pipeline. this would lead to immediate and clear error messages when the user is at fault. (inspired by design by contract.)

[azat-badretdin]

I also noticed that characters such as 'í' lead to crashes when used in names. (However, 'ä' in street seems to work.)

Could you please two corresponding submol.yaml files for these cases?

[azat-badretdin]

Meanwhile I tried to add i-acute and a-umlaut to submol.yaml and both cases failed at the step yaml2json.py

That py script is suspect number one now.

[TheBigFatTony]

Could you please two corresponding submol.yaml files for these cases?

í in names -> failure:

the error message is:

(...)
    and not(contains(@code, "SEQ_PKG_ComponentMissingTitle")) 
    and not(contains(@code, "SEQ_DESCR_ChromosomeLocation")) 
    and not(contains(@code, "SEQ_DESCR_MissingLineage")) 
    and not(contains(@code, "SEQ_DESCR_NoTaxonID")) 
    and not(contains(@code, "SEQ_FEAT_ShortIntron")) 
]
'
Failer nodes:
<?xml version="1.0" encoding="UTF-8"?>
<message severity="ERROR" seq-id="lcl|L43967.2" code="SEQ_FEAT_BadCharInAuthorLastName">Bad characters in author Sch##nb##chler</message>

[2019-11-19 14:51:09] WARNING [job Prepare_Unannotated_Sequences_asnvalidate_evaluate] completed permanentFail
[2019-11-19 14:51:09] WARNING [step Prepare_Unannotated_Sequences_asnvalidate_evaluate] completed permanentFail
[2019-11-19 14:51:09] INFO [workflow standard_pgap] completed permanentFail
[2019-11-19 14:51:09] WARNING [step standard_pgap] completed permanentFail
[2019-11-19 14:51:09] INFO [workflow ] completed permanentFail
[2019-11-19 14:51:09] WARNING Final process status is permanentFail
{
    "gbk": null,
    "gff": null,
(...)

this is the corresponding submol.yaml:

topology: circular
comment: 'There is no really a biologist Arnold Schwarzenegger'
consortium: 'SkyNet consortium'
sra:
    - accession: 'ERR2193926'
tp_assembly: true
organism:
    genus_species: 'Mycoplasma genitalium' 
    strain: 'replaceme'
contact_info:
    last_name: 'Schönbächler'
    first_name: 'Jane'
    email: '[email protected]'
    organization: 'Institute of Klebsiella foobarensis research'
    department: 'Department of Using NCBI'
    phone: '301-555-0245'
    street: '1234 Main St'
    city: 'Docker'
    postal_code: '12345'
    country: 'Lappland'
    
authors:
    -     author:
            first_name: 'Arnold'
            last_name: 'Schwarzenegger'
    -     author:
            first_name: 'Linda'
            last_name: 'Hamilton'
bioproject: 'PRJNA9999999'
biosample: 'SAMN99999999'      
# -- Locus tag prefix - optional. Limited to 9 letters. Unless the locus tag prefix was officially assigned by NCBI, ENA, or DDBJ, it will be replaced upon submission of the annotation to NCBI and is therefore temporary and not to be used in publications. If not provided, pgaptmp will be used.
locus_tag_prefix: 'tmp'
publications:
    - publication:
        pmid: 16397293
        title: 'Discrete CHARMm of Klebsiella foobarensis. Journal of Improbable Results, vol. 34, issue 13, pages: 10001-100005, 2018'
        status: published  # this is enum: controlled vocabulary
        authors:
            - author:
                first_name: 'Arnold'
                last_name: 'Schwarzenegger'
            - author:
                  first_name: 'Linda'
                  last_name: 'Hamilton'

umlauts in street and author names -> success:

topology: circular
comment: 'There is no really a biologist Arnold Schwarzenegger'
consortium: 'SkyNet consortium'
sra:
    - accession: 'ERR2193926'
tp_assembly: true
organism:
    genus_species: 'Mycoplasma genitalium' 
    strain: 'replaceme'
contact_info:
    last_name: 'Hamilton'
    first_name: 'Jane'
    email: '[email protected]'
    organization: 'Institute of Klebsiella foobarensis research'
    department: 'Department of Using NCBI'
    phone: '301-555-0245'
    street: '1234 Mäín Ströüt'
    city: 'Docker'
    postal_code: '12345'
    country: 'Lappland'
    
authors:
    -     author:
            first_name: 'Arnold'
            last_name: 'Schwarzenegger'
    -     author:
            first_name: 'Linda'
            last_name: 'Hamilton'
bioproject: 'PRJNA9999999'
biosample: 'SAMN99999999'      
# -- Locus tag prefix - optional. Limited to 9 letters. Unless the locus tag prefix was officially assigned by NCBI, ENA, or DDBJ, it will be replaced upon submission of the annotation to NCBI and is therefore temporary and not to be used in publications. If not provided, pgaptmp will be used.
locus_tag_prefix: 'tmp'
publications:
    - publication:
        pmid: 16397293
        title: 'Discrete CHARMm of Klebsiella foobarensis. Journal of Improbable Results, vol. 34, issue 13, pages: 10001-100005, 2018'
        status: published  # this is enum: controlled vocabulary
        authors:
            - author:
                first_name: 'Arnold'
                last_name: 'Schönbächler'
            - author:
                  first_name: 'Linda'
                  last_name: 'Hamilton'

[azat-badretdin]

Thanks! That is very helpful. I see now that in your environment this case magically advances further than in my environment (where it breaks at yaml2json converter). I added European characters to our TeamCity case and we should be able to push this through soon.