the nmdc-schema may contain some Biosample slots that belong exclusively to the submission-schema

[turbomam]

For example, rna_volume is defined in nmdc-schema and associated with Biosample. We even have some tests for it.

If it is important to capture that information in the SubmissionPortal, so that it can be passed on to user facilities, but it is never going to be saved into MongoDB, maybe it doesn't belong in nmdc-schema?

Right now, I don't think we have a way to introduce slots into submission-schema other than by extraction from nmdc-schema, but that doesn't seem like a big technical challenge.

see this example data file, which I revised in a currently-unmerged PR.

[mslarae13]

below is a list of slots that are in nmdc schema, but are only needed in the submission portal. These slots need to be store on the submission portal side, and able to be exported for the JGI metadata template. But do not need captured by NMDC

<style type="text/css"></style>

emsl_store_temp	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/emsl.yaml	EMSL	Note
sample_shipped	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/emsl.yaml	EMSL
sample_type	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/emsl.yaml	EMSL
dna_absorb1	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	IF we keep it, it should map to processed sample, not biosamples
dna_absorb2	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	IF we keep it, it should map to processed sample, not biosamples
dna_collect_site	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	I think we can remove this slot & just put the extension that is selected. so 'soil' but a JGI team member should weigh in
dna_cont_type	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI
dna_cont_well	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI
dna_container_id	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI
dna_dnase	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI
dna_organisms	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	Remove completely?
dna_project_contact	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	Will need for JGI API
dna_samp_id	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	this is effectively an alterntaive identifier for a processed sample which is the output of Class Extraction. JGI doesn't really expose these identifiers externally so low priority to keep. Would be worth keeping in the submission portal postgres in case we ever need to pul this in.
dna_sample_format	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	not informative for workflows
dna_sample_name	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	Would map to processed sample if we do decided to keep it
dna_seq_project_pi	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	Will need for JGI API, principal_investigator will be available on OmicsProcessing/DataGeneration
dna_seq_project_name	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	Will pull from JGI API
dna_volume	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI
proposal_dna	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metagenomics.yaml	JGI	Will pull from JGI API
dna_concentration	nmdc-schema/src/schema/basic_slots.yaml Line 84 in df7bd5e dna_concentration:	JGI	Would map to processed sample if we do decided to keep it. Why is it in basic_slots and not jgi_metagenomics.yaml
proposal_rna	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Will pull from JGI API
rna_absorb1	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	IF we keep it, it should map to processed sample, not biosamples
rna_absorb2	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	IF we keep it, it should map to processed sample, not biosamples
rna_collect_site	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	I think we can remove this slot & just put the extension that is selected. so 'soil' but a JGI team member should weigh in
rna_concentration	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Would map to processed sample if we do decided to keep it. Why is it in basic_slots and not jgi_metagenomics.yaml
rna_cont_type	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI
rna_cont_well	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI
rna_container_id	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI
rna_organisms	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Remove completely?
rna_project_contact	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Will need for JGI API
rna_samp_id	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	this is effectively an alterntaive identifier for a processed sample which is the output of Class Extraction. JGI doesn't really expose these identifiers externally so low priority to keep. Would be worth keeping in the submission portal postgres in case we ever need to pul this in.
rna_sample_format	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	not informative for workflows
rna_sample_name	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Would map to processed sample if we do decided to keep it
rna_seq_project_pi	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Will need for JGI API, principal_investigator will be available on OmicsProcessing/DataGeneration
rna_seq_project_name	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI	Will pull from JGI API
rna_volume	https://github.com/microbiomedata/nmdc-schema/blob/main/src/schema/portal/jgi_metatranscriptomics.yaml	JGI

[mslarae13]

@pkalita-lbl @turbomam See the table above.

Next step for these JGI and EMSL slots I propose

• Remove from the linked .yaml files
• Add to NMDC submission schema. @pkalita-lbl Can you add some direction on how I can do this (what file and where to add/move these slots to)

[pkalita-lbl]

The slots would need to be defined in this TSV file: https://github.com/microbiomedata/submission-schema/blob/main/schemasheets/tsv_in/slots.tsv

The association of slots to classes happens in this TSV file: https://github.com/microbiomedata/submission-schema/blob/main/schemasheets/tsv_in/classes.tsv. You can see that a lot of the slots in question are already associated with the appropriate Interface classes. Those would just need to be reviewed to make sure they're still accurate.

[mslarae13]

From slack

@turbomam "If some UF slots are gong to be moved to submission-schema in the near future, then I would prefer not to move them around within the nmdc-schema modules. I think Alicia and I moved some slots because she was using them outside of their original UF use-case, and that kept the schema from building. We should have included you ion that decison.
I'm working on an issue to make all modules self sufficient (ie build on their own), and this will address the question you have asked along the line "how do I know which module to put new content in?" Can we leave the UF slots where they are until then?"

In summary
for user facility slots we want to keep in NMDC schema, I'll add alias. In a later task we'll complete this issue, getting the user facility slots that NMDC does NOT need to track removed from NMDC schema, and only have them in NMDC submission portal. We will also later decide if the slots that we DO capture should remain in separate .yaml files or be moved to basic_slots.

[ssarrafan]

@mslarae13 @pkalita-lbl is this an active issue? I'm going to remove this from this sprint and add to the backlog but if it's active please add to a future sprint.

[mslarae13]

See also microbiomedata/metadata-for-user-facility-template-transformations#16

[mslarae13]

This is blocked until we resolve how. Converting to a discussion.