Submit EcoFAB data to NCBI

[aclum]

Deliverable this task is associated with

See Deliverables tab here:

• [Add the Deliverable #]

RACI

Tag people in their roles

• Responsible: @sujaypatil96, @chienchi
• Accountable: @turbomam
• Consulted: @aclum, @chienchi
• Informed: @mslarae13

Describe the the task

• [ ] Generate biosample and SRA submission xml for EcoFab submission from Vlastimil Novak

Criteria for completion

• ncbi biosample and SRA identifiers

Estimate people time

• 2, 20 hours

Completion Date (Goal)

• Aug 16

Target Sprint Start & End Dates

• Start: July 29
• End: Aug 22

Tag Blocker/Contingent upon issues

• blocked on submission portal to mongo ingest @mslarae13 please add the ticket for that once it exists.

cc @ssarrafan

[aclum]

For the SRA block use library_strategy = AMPLICON library_source = METAGENOMIC library_selection = PCR library_layout = paired platform = ILLUMINA instrument_model = Illumina MiSeq

[ssarrafan]

Blocked by https://github.com/microbiomedata/issues/issues/776 Montana is reviewing Mongo ingest today

[aclum]

This data is in mongo prod now see https://github.com/microbiomedata/issues/issues/812 Study ID is nmdc:sty-11-ev70y104 The data to make the sra records is here /global/cfs/cdirs/m3408/gsharing/ecofab/ecofab_data

@sujaypatil96 this is ready to start work on.

[sujaypatil96]

Starting work on this today, 8/7.

[sujaypatil96]

This needs to be moved to the next sprint.

[sujaypatil96]

Status update:

• Currently (as of this message) we do not have an existing BioProject record/id for the EcoFAB study on NCBI yet, so we will need to create one. In my pipeline I am using a placeholder BioProject id (PRJNA123456) so if you see that id in the XML know that it will be replaced with the correct NCBI BioProject id once we’ve created it
• Organization name we are using is “Lawrence Berkeley National Laboratory” and “broker name” is “National Microbiome Data Collaborative”. Please let me know if either of those needs to be corrected/changed to something else In the SRA <Action> blocks:
  • We have nested <Identifier> tags. For the NEON studies we were populating that field with information from LibraryPreparation.protocol_link.name but unfortunately we can’t do that for records in this study, so I am simply populating it with Biosample.name == Biosample.samp_name == “sample name” from spreadsheet
• We are also missing Library attribute information for library_construction_protocol in this comment
• We were using env_package to infer the value for <Package> in the BioSample <Action>blocks which makes it look like this: <Package>MIMS.me.miscellaneous natural or artificial environment.6.0</Package> Ideally, we would want to use a standard NCBI package name from here: https://www.ncbi.nlm.nih.gov/biosample/docs/packages/
• We can see 140 <Action> blocks for BioSample records and 90 blocks for SRA records which are both the expected number of blocks

Refer to the squad slack channel for a preview of the XML export file.

ETA: ideally, end of the week/sprint.

[aclum]

Actively in progress but we are still iterating over the code that generates the xml for this new package so moving the next sprint.

[aclum]

We tried to submit this as a test but are having trouble with the bioproject IDs.

[ssarrafan]

We tried to submit this as a test but are having trouble with the bioproject IDs.

@sujaypatil96 any update on the bioproject IDs issue?

[chienchi]

The test submission successfully and will submit to production. Move to next sprint and hope it will be done early next week.

[chienchi]

The submission to NCBI went through.