nmdc-data

name	nmdc-data
description	Skills for querying microbiome data in NMDC. Use this proactively for anything related to microbes and microbiomes. Use for anything involving NMDC IDs (e.g. prefixed `nmdc:`)

Command line tool

These tools require nmdc on the command line.

Use nmdc --help or nmdc COMMAND --help for more detailed info, the most common operations are shown here.

Most tools allow export to CSV, by appending options:

nmdc COMMAND MAIN_OPTS -f csv -o out.csv

Looking up and searching biosamples

Lookup by id:

nmdc biosample --id nmdc:bsm-11-NNNNNN

Finding samples for a study study:

nmdc biosample --filter 'associated_studies: nmdc:sty-11-NNNNNN'

Searching in a bounding box:

nmdc biosample --bbox "30,-120,40,-100"

Search by GOLD classification:

nmdc biosample --filter 'ecosystem: Engineered'

If your starting point is a different kind of object, use link to navigate.

Studies

Fetch a study by ID:

nmdc study --id nmdc:sty-11-34xj1150

Search by ecosystem:

nmdc study --filter 'ecosystem_category: Aquatic' --limit 5

Finding linked objects

The NMDC data model at a high level is:

• Study -> Biosample -> DataObject

You can traverse objects up and down and across multiple hops using link

Example:

nmdc link nmdc:dobj-11-m71bzv60

Filtering by object type:

nmdc link nmdc:dobj-11-m71bzv60 -t nmdc:Biosample

Note that the link command will also include workflow objects but these typically don't need to be referenced individually

Finding data for a study, sample, or data object

If you have a data object ID, then you can fetch its metadata:

nmdc data-object --id nmdc:dobj-11-m71bzv60

The JSON will include a URL. Download this the usual way

Dumping data for a study

If you need to dump multiple files for analysis, use dump-study

Example:

nmdc dump-study nmdc:sty-11-aygzgv51 ./my_data --download-data -t "Kraken2 Taxonomic Classification, Functional Annotation GFF"

This creates a structure, with the study at the top, and all data organized under a folder for each sample:

my_data/
  study.json -- metadata about nmdc:sty-11-aygzgv51
  nmdc_bsm-11-FIRST-SAMPLE
    biosample.json -- metadata about the first sample
    data_objects/
      nmdc_wfgan-11-NNNNNN_functional_annotation.gff -- genes with functional ann in col9. can be analyzed using GFF skill
      nmdc_wfrbt-11-NNNNNN_kraken2_classification.tsv -- taxonomic classification.