corpus: feature-table-corpus
generated: '2026-09-21'
description: Real, traceable examples of GFF and other genome feature table content, for testing and for
  modeling work on a unified feature schema.
tiers:
  vendored: the file is in this repository, with its origin URL and a checksum
  linked: the file is too large or not redistributable; a stable public URL is recorded instead
  restricted: the content sits behind a login; recorded for completeness and not fetchable here
  derived: built here from a vendored file, or from a restricted one whose md5 and redistribution_basis are recorded (docs/restricted-sources.md), by exactly one documented change; traceable, but NOT observed
    in the wild. See the validity field, which says whether the result is invalid GFF3 or valid and merely
    refused
  not located: known to exist, no public URL found; recorded so the gap stays visible
entry_count: 116
entries:
- id: nmdc-cath-funfam
  label: CATH FunFams (Functional Families) Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-ghk5e506.1_cath_funfam.gff
  format: GFF3
  source: NMDC
  producer: CATH functional families database search
  producer_kind: database search
  bytes: 295
  md5: 385ae451b6c19e33abcf3c37f0ae4b69
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-4e682h13/nmdc:wfmgan-11-ghk5e506.1/nmdc_wfmgan-11-ghk5e506.1_cath_funfam.gff
  origin_record: nmdc:dobj-11-3pw65847
  generated_by: null
  selection: smallest of 599 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-crt
  label: CRT Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-bvg4py20.1_crt.gff
  format: GFF3
  source: NMDC
  producer: CRISPR Recognition Tool, feature finder
  producer_kind: feature finder
  bytes: 688
  md5: e5b1788663cc6c0c7b50a48558cb49bb
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:manif-15-2dqj6n85/nmdc:wfmgan-11-bvg4py20.1/nmdc_wfmgan-11-bvg4py20.1_crt.gff
  origin_record: nmdc:dobj-11-3cjcxz28
  generated_by: nmdc:wfmgan-11-bvg4py20.1
  selection: smallest of 507 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-cog
  label: Clusters of Orthologous Groups (COG) Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-ghk5e506.1_cog.gff
  format: GFF3
  source: NMDC
  producer: COG database search
  producer_kind: database search
  bytes: 268
  md5: 9ddb5396a789d11cf01370ebb8daf479
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-4e682h13/nmdc:wfmgan-11-ghk5e506.1/nmdc_wfmgan-11-ghk5e506.1_cog.gff
  origin_record: nmdc:dobj-11-22ky0702
  generated_by: null
  selection: smallest of 599 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-functional-annotation
  label: Functional Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-hrn8ep39.1_functional_annotation.gff
  format: GFF3
  source: NMDC
  producer: JGI IMG annotation pipeline, functional roll-up
  producer_kind: roll-up
  bytes: 246
  md5: 778d966552b57f82ece65e6581ce574e
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-q17q2c37/nmdc:wfmgan-11-hrn8ep39.1/nmdc_wfmgan-11-hrn8ep39.1_functional_annotation.gff
  origin_record: nmdc:dobj-11-16876g28
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-genemark
  label: Genemark Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-prsstp65.1_genemark.gff
  format: GFF3
  source: NMDC
  producer: GeneMark, gene caller
  producer_kind: gene caller
  bytes: 3333
  md5: 5121dc1269540aeb5f6b9a9a7dcd914f
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-12-vanwrb64/nmdc:wfmgan-11-prsstp65.1/nmdc_wfmgan-11-prsstp65.1_genemark.gff
  origin_record: nmdc:dobj-11-0t0kn065
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-ko-ec
  label: KO_EC Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-v44mx103.1_ko_ec.gff
  format: GFF3
  source: NMDC
  producer: KEGG Orthology and Enzyme Commission assignment
  producer_kind: database search
  bytes: 289
  md5: 7751226802a1cc5db22c26d5c8956938
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-xpbh2e04/nmdc:wfmgan-11-v44mx103.1/nmdc_wfmgan-11-v44mx103.1_ko_ec.gff
  origin_record: nmdc:dobj-11-30x39t60
  generated_by: null
  selection: smallest of 599 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-pfam
  label: Pfam Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-tn3gnf41.1_pfam.gff
  format: GFF3
  source: NMDC
  producer: Pfam database search
  producer_kind: database search
  bytes: 4626
  md5: 7fdd0333e764c69d87e95094713abf81
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-g54wdz69/nmdc:wfmgan-11-tn3gnf41.1/nmdc_wfmgan-11-tn3gnf41.1_pfam.gff
  origin_record: nmdc:dobj-11-2cbp5z37
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-prodigal
  label: Prodigal Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-9ya9xh30.1_prodigal.gff
  format: GFF3
  source: NMDC
  producer: Prodigal, gene caller
  producer_kind: gene caller
  bytes: 2033
  md5: 92d78177f3c3d0c410d0b57c94ebf22c
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-cfaemn69/nmdc:wfmgan-11-9ya9xh30.1/nmdc_wfmgan-11-9ya9xh30.1_prodigal.gff
  origin_record: nmdc:dobj-11-2jq65c25
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-rfam
  label: RFAM Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-p2fy8323.1_rfam.gff
  format: GFF3
  source: NMDC
  producer: Rfam, feature finder
  producer_kind: feature finder
  bytes: 735
  md5: cd367ad9b2445ba7a40049a2fc052739
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-02tp7916/nmdc:wfmgan-11-p2fy8323.1/nmdc_wfmgan-11-p2fy8323.1_rfam.gff
  origin_record: nmdc:dobj-11-26sjcj43
  generated_by: null
  selection: smallest of 597 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-smart
  label: SMART Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-1sz9qv06.1_smart.gff
  format: GFF3
  source: NMDC
  producer: SMART database search
  producer_kind: database search
  bytes: 265
  md5: 8107e708d5abe55ab2c25fb45c39cf8e
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:dgns-11-dy2n0z32/nmdc:wfmgan-11-1sz9qv06.1/nmdc_wfmgan-11-1sz9qv06.1_smart.gff
  origin_record: nmdc:dobj-11-03kh9653
  generated_by: nmdc:wfmgan-11-1sz9qv06.1
  selection: smallest of 598 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-superfam
  label: SUPERFam Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-5pdxej61.1_supfam.gff
  format: GFF3
  source: NMDC
  producer: SUPERFAMILY database search
  producer_kind: database search
  bytes: 603
  md5: 6460a86d5eab227d644578d0c15bea88
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-12-fmd4p583/nmdc:wfmgan-11-5pdxej61.1/nmdc_wfmgan-11-5pdxej61.1_supfam.gff
  origin_record: nmdc:dobj-11-10126e89
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-structural-annotation
  label: Structural Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-hrn8ep39.1_structural_annotation.gff
  format: GFF3
  source: NMDC
  producer: JGI IMG annotation pipeline, structural roll-up
  producer_kind: roll-up
  bytes: 158
  md5: 4b8627ca236a7d9753185ccdfd6baada
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-q17q2c37/nmdc:wfmgan-11-hrn8ep39.1/nmdc_wfmgan-11-hrn8ep39.1_structural_annotation.gff
  origin_record: nmdc:dobj-11-25eqc554
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-tigrfam
  label: TIGRFam Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-8qy5m271.1_tigrfam.gff
  format: GFF3
  source: NMDC
  producer: TIGRFAM database search
  producer_kind: database search
  bytes: 191
  md5: b6b168f24def6d0d12f238d31cf48f35
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-haha6973/nmdc:wfmgan-11-8qy5m271.1/nmdc_wfmgan-11-8qy5m271.1_tigrfam.gff
  origin_record: nmdc:dobj-11-2j11wr22
  generated_by: nmdc:wfmgan-11-8qy5m271.1
  selection: smallest of 595 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-trna
  label: TRNA Annotation GFF
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-p2fy8323.1_trna.gff
  format: GFF3
  source: NMDC
  producer: tRNAscan, feature finder
  producer_kind: feature finder
  bytes: 183
  md5: 28ee4b6dda9c66f3c6cf755931725e35
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-02tp7916/nmdc:wfmgan-11-p2fy8323.1/nmdc_wfmgan-11-p2fy8323.1_trna.gff
  origin_record: nmdc:dobj-11-24g8zk65
  generated_by: null
  selection: smallest of 595 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-11'
  license: CC BY 4.0
- id: nmdc-ec
  label: Annotation Enzyme Commission
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-6kajck87.1_ec.tsv
  format: TSV
  source: NMDC
  producer: EC annotation, database search
  producer_kind: database search
  bytes: 92
  md5: 7ff938626fae991f42911f5e19f8d703
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-azzp1136/nmdc:wfmgan-11-6kajck87.1/nmdc_wfmgan-11-6kajck87.1_ec.tsv
  origin_record: nmdc:dobj-11-3btx1w33
  generated_by: null
  selection: smallest of 599 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-17'
  license: CC BY 4.0
  notes: Not GFF. A tab-delimited, one-row-per-feature table with no header row, 11 fields, no
    column labels anywhere in the file. Field 1 is a feature id matching the GFF column-9 ID pattern
    from the same workflow run; field 3 is an EC CURIE; fields 4 and 9 are floats shaped like a percent
    (0-100) and a small exponential number respectively, consistent with a BLAST/HMMER-style hit
    record, but that is a shape observation, not a confirmed column meaning, since there is no header
    and no IMG/JGI pipeline documentation for this exact file has been checked yet. Column meanings are
    unverified
    and should not be asserted without reading the producing tool's own docs or source. This is one of
    the two largest tables in BERDL's nmdc.results namespace at production scale
    (annotation_enzyme_commission, 1.23 billion rows per project-nmdc-lakehouse-dump-and-reload), so
    its schema matters more than its 92-byte sample suggests.
- id: nmdc-ko
  label: Annotation KEGG Orthology
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-vc2g4r14.1_ko.tsv
  format: TSV
  source: NMDC
  producer: KO annotation, database search
  producer_kind: database search
  bytes: 542
  md5: 986fa951a1ca5923a0ee297d71142b05
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-nz0hh615/nmdc:wfmgan-11-vc2g4r14.1/nmdc_wfmgan-11-vc2g4r14.1_ko.tsv
  origin_record: nmdc:dobj-11-0s5bnb64
  generated_by: null
  selection: smallest of 600 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-17'
  license: CC BY 4.0
  notes: Not GFF. Same 11-field, no-header shape as nmdc-ec, field 3 a KO CURIE in place of EC. Same
    caveat applies, the numeric fields' meanings are unverified. The other of the two largest BERDL
    nmdc.results tables (annotation_kegg_orthology, 1.83 billion rows per
    project-nmdc-lakehouse-dump-and-reload).
- id: nmdc-crispr-terms
  label: Crispr Terms
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-vypv6e87.1_crt.crisprs
  format: CRT
  source: NMDC
  producer: CRT (CRISPR Recognition Tool)
  producer_kind: feature finder
  bytes: 236
  md5: 7f0b1a9e4635ef2facfd8aa071a73796
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-bnhaje30/nmdc:wfmgan-11-vypv6e87.1/nmdc_wfmgan-11-vypv6e87.1_crt.crisprs
  origin_record: nmdc:dobj-11-0bvfgb75
  generated_by: null
  selection: smallest of 505 records of this data_object_type sampled from the NMDC public API
  retrieved: '2026-09-17'
  license: CC BY 4.0
  notes: Not GFF, and a different field count from nmdc-ec/nmdc-ko (6 fields here, not 11). No header.
    Field 1 is a sequence id, field 2 is the constant value 1 across all three rows in this sample,
    field 3 increases per row (106, 154, 199, plausibly a position but unverified), field 4 is a
    20-21bp sequence, field 5 is a variable-length sequence that is empty on the last row of the array,
    and field 6 is the constant letter c on all three rows. Column names are unverified against any
    CRT documentation; this is CRT's (CRISPR Recognition Tool) own output format and is not
    standardized anywhere else in this corpus.
- id: ncbi-refseq-GCF_000819615-1
  label: NCBI RefSeq GFF3, Escherichia phage phiX174
  tier: vendored
  path: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 6571
  md5: 704a57e8aeb9e7c80930a88fad0b1994
  origin_url: https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/819/615/GCF_000819615.1_ViralProj14015/GCF_000819615.1_ViralProj14015_genomic.gff.gz
  origin_record: GCF_000819615.1
  selection: chosen as a complete, tiny reference annotation from the canonical GFF3 producer
  retrieved: '2026-09-11'
  license: public domain (US Government work)
  vendored_form: decompressed from the gzip the origin URL serves, because this repository does not carry
    binary files
- id: ncbi-refseq-GCF_000840245-1
  label: NCBI RefSeq GFF3, Escherichia phage lambda
  tier: vendored
  path: corpus/sources/ncbi-refseq/ncbi_refseq_lambda_GCF_000840245.1.gff
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 58152
  md5: e1c633cadbff78104fae79ab1bee4c0c
  origin_url: https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/840/245/GCF_000840245.1_ViralProj14204/GCF_000840245.1_ViralProj14204_genomic.gff.gz
  origin_record: GCF_000840245.1
  selection: chosen as a complete, tiny reference annotation from the canonical GFF3 producer
  retrieved: '2026-09-11'
  license: public domain (US Government work)
  vendored_form: decompressed from the gzip the origin URL serves, because this repository does not carry
    binary files
- id: agat-flavor-suite
  label: AGAT flavor and syntax test suite
  tier: linked
  format: GFF, GFF1, GFF2, GFF2.5, GFF3, GTF, GTF2, GTF2.1, GTF2.2, GTF2.5, GTF3
  source: NBIS Sweden
  producer: hand-built test fixtures
  producer_kind: test suite
  origin_url: https://github.com/NBISweden/AGAT/tree/master/t
  license: GPL-3.0
  why_not_vendored: GPL-3.0 is copyleft; linking avoids putting a license obligation on this corpus
  notes: The single best source for the named format flavors. AGAT's own documentation claims more than
    30 distinct cases. Subdirectories t/gff_version and t/gff_syntax are the relevant ones.
  retrieved: '2026-09-11'
- id: genometools-testdata
  label: GenomeTools GFF3 test data
  tier: linked
  format: GFF3
  source: GenomeTools
  producer: hand-built test fixtures
  producer_kind: test suite
  origin_url: https://github.com/genometools/genometools/tree/master/testdata
  license: unstated in repository metadata
  why_not_vendored: license could not be determined; 315 GFF files matched a name filter
  notes: Largest single pile of GFF3 edge cases found. Includes malformed files used to test the validator.
  retrieved: '2026-09-11'
- id: gff3toolkit-examples
  label: GFF3toolkit example files
  tier: linked
  format: GFF3
  source: USDA i5K
  producer: arthropod genome projects
  producer_kind: curated real annotation
  origin_url: https://github.com/NAL-i5K/GFF3toolkit/tree/master/example_file
  license: unstated in repository metadata
  why_not_vendored: license could not be determined
  notes: Real arthropod annotations with the merge and quality-control problems the AgBioData group describes.
  retrieved: '2026-09-11'
- id: so-gff3-spec
  label: Sequence Ontology GFF3 specification, with canonical examples
  tier: linked
  format: GFF3
  source: Sequence Ontology
  producer: specification authors
  producer_kind: specification
  origin_url: https://github.com/The-Sequence-Ontology/Specifications/blob/master/gff3.md
  license: no license file in repository
  why_not_vendored: no license; and the examples are worth reading in their specification context
  notes: Normative. Defines the nine columns, the parent and child model, percent-encoding in column 9,
    and the rule that start <= end on every feature. Its worked examples are the reference for canonical
    structure.
  retrieved: '2026-09-11'
  kind: specification
- id: nmdc-functional-annotation-gff-large
  label: NMDC Functional Annotation GFF, full size example
  tier: linked
  format: GFF3
  source: NMDC
  producer: JGI IMG annotation pipeline, functional roll-up
  producer_kind: roll-up
  origin_url: https://api.microbiomedata.org/nmdcschema/data_object_set
  license: CC BY 4.0
  why_not_vendored: the collection is about 4,836 files and 1,309 GB, averaging 277 MB each
  notes: Column 9 carries every annotation system at once, with multivalued pfam, tigrfam, cath_funfam
    and superfamily keys. Query the API by data_object_type to list them. Tracked in https://github.com/microbiomedata/nmdc-lakehouse/issues/84
  retrieved: '2026-09-11'
- id: gencode-human-gtf
  label: GENCODE human annotation, GTF
  tier: linked
  format: GTF
  source: GENCODE at EMBL-EBI
  producer: GENCODE
  producer_kind: reference annotation
  origin_url: https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/latest_release/
  license: free for academic and commercial use per GENCODE terms
  why_not_vendored: tens of megabytes compressed
  notes: The GTF lineage in its most widely consumed form. Directory URL rather than a file URL, because
    the release number moves.
  retrieved: '2026-09-11'
- id: jgi-phytozome
  label: JGI Phytozome plant annotations
  tier: restricted
  format: GFF3
  source: JGI
  producer: JGI plant annotation pipeline
  producer_kind: reference annotation
  origin_url: https://phytozome-next.jgi.doe.gov/
  license: JGI data use policy
  barrier: requires a JGI account and accepting a data use policy
  notes: A BER producer of plant feature tables. Recorded because a unified BER model has to cover it,
    not because it can be fetched here.
  retrieved: '2026-09-11'
- id: jgi-mycocosm
  label: JGI MycoCosm fungal annotations
  tier: restricted
  format: GFF3
  source: JGI
  producer: JGI fungal annotation pipeline
  producer_kind: reference annotation
  origin_url: https://mycocosm.jgi.doe.gov/
  license: JGI data use policy
  barrier: requires a JGI account
  notes: Same reasoning as Phytozome.
  retrieved: '2026-09-11'
- id: jgi-img
  label: JGI IMG annotations
  tier: restricted
  format: GFF3
  source: JGI
  producer: IMG annotation pipeline
  producer_kind: reference annotation
  origin_url: https://img.jgi.doe.gov/
  license: JGI data use policy
  barrier: requires a JGI account for bulk access
  notes: This is the pipeline that produced every vendored NMDC file in this corpus. The NMDC copies are
    the open route to the same output. Two public isolate records from the JGI Data Portal are vendored
    separately under corpus/sources/jgi-img/.
  retrieved: '2026-09-11'
- id: ncbi-refseq-gtf-GCF_000819615-1
  label: NCBI RefSeq GTF, Escherichia phage phiX174
  tier: vendored
  path: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gtf
  format: GTF
  gtf_version: '2.2 (declared in the file''s own #gtf-version pragma)'
  source: NCBI RefSeq
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 14316
  md5: 83a7c266f67a91d4ac51a4e9b39babab
  origin_url: https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/819/615/GCF_000819615.1_ViralProj14015/GCF_000819615.1_ViralProj14015_genomic.gtf.gz
  origin_record: GCF_000819615.1
  selection: the GTF NCBI publishes beside the GFF3 already in this corpus, so the same annotation is
    available in both formats
  vendored_form: decompressed from the gzip the origin URL serves
  retrieved: '2026-09-11'
  license: public domain (US Government work)
  notes: Pairs with the GFF3 of the same accession. Useful for testing that a model round-trips between
    the two column-9 grammars.
- id: ncbi-refseq-gtf-GCF_000840245-1
  label: NCBI RefSeq GTF, Escherichia phage lambda
  tier: vendored
  path: corpus/sources/ncbi-refseq/ncbi_refseq_lambda_GCF_000840245.1.gtf
  format: GTF
  gtf_version: '2.2 (declared in the file''s own #gtf-version pragma)'
  source: NCBI RefSeq
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 96276
  md5: 149c9381a2267c837dee92f12c191859
  origin_url: https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/840/245/GCF_000840245.1_ViralProj14204/GCF_000840245.1_ViralProj14204_genomic.gtf.gz
  origin_record: GCF_000840245.1
  selection: the GTF NCBI publishes beside the GFF3 already in this corpus, so the same annotation is
    available in both formats
  vendored_form: decompressed from the gzip the origin URL serves
  retrieved: '2026-09-11'
  license: public domain (US Government work)
  notes: Pairs with the GFF3 of the same accession. Useful for testing that a model round-trips between
    the two column-9 grammars.
- id: spec-insdc-feature-table
  label: INSDC Feature Table Definition
  tier: linked
  kind: specification
  origin_url: https://www.insdc.org/submitting-standards/feature-table/
  source: INSDC
  license: not stated
  notes: The ancestor of all of this. GenBank, EMBL and DDBJ feature keys and qualifiers. Predates GFF
    and still governs what a 'feature' means in sequence databases.
  retrieved: '2026-09-11'
- id: spec-ddbj-flat-file
  label: DDBJ flat file and feature table documentation
  tier: linked
  kind: specification
  origin_url: https://www.ddbj.nig.ac.jp/ddbj/flat-file-e.html
  source: DDBJ
  license: not stated
  notes: The most readable rendering of the INSDC feature table.
  retrieved: '2026-09-11'
- id: spec-gvf-so
  label: GVF specification, Sequence Ontology
  tier: linked
  kind: specification
  origin_url: https://github.com/The-Sequence-Ontology/Specifications/blob/master/gvf.md
  source: Sequence Ontology
  license: no license file in repository
  notes: GFF3 with variant-specific attributes. Shows how the format is meant to be extended rather than
    forked.
  retrieved: '2026-09-11'
- id: spec-gtf22
  label: GTF 2.2 specification
  tier: linked
  kind: specification
  origin_url: http://mblab.wustl.edu/GTF22.html
  source: Brent Lab, Washington University
  license: not stated
  notes: The version the vendored NCBI GTF files declare. The clearest statement of how GTF's column 9
    differs from GFF3's.
  retrieved: '2026-09-11'
- id: spec-gff2-gmod
  label: GFF2 description, GMOD
  tier: linked
  kind: specification
  origin_url: http://gmod.org/wiki/GFF2
  source: GMOD
  license: not stated
  notes: The Sanger-lineage format Chado and older tools consume. Documents why GFF2 could not express
    parent and child properly, which is why GFF3 exists.
  retrieved: '2026-09-11'
- id: spec-gff3-gmod
  label: GFF3 description, GMOD
  tier: linked
  kind: specification
  origin_url: http://gmod.org/wiki/GFF3
  source: GMOD
  license: not stated
  notes: Practical companion to the normative specification.
  retrieved: '2026-09-11'
- id: spec-bed-hts
  label: BED specification version 1
  tier: linked
  kind: specification
  origin_url: https://samtools.github.io/hts-specs/BEDv1.pdf
  source: Samtools and GA4GH
  license: no license file in repository
  why_not_vendored: PDF, and this repository carries no binary files
  notes: Other feature-table content. BED is the minimal interval format and the one most feature tables
    get reduced to for browsing.
  retrieved: '2026-09-11'
- id: spec-ucsc-formats
  label: UCSC genome browser format reference
  tier: linked
  kind: specification
  origin_url: https://genome.ucsc.edu/FAQ/FAQformat.html
  source: UCSC
  license: not stated
  notes: BED, bigBed, genePred, PSL, narrowPeak and GFF/GTF as UCSC consumes them. The broadest single
    list of real feature-table formats.
  retrieved: '2026-09-11'
- id: spec-ncbi-gff3-guide
  label: NCBI guidance on GFF3 and GTF
  tier: linked
  kind: specification
  origin_url: https://www.ncbi.nlm.nih.gov/genbank/genomes_gff/
  source: NCBI
  license: public domain (US Government work)
  notes: What the producer of the vendored NCBI files says it emits, including its deviations.
  retrieved: '2026-09-11'
- id: spec-chado-feature
  label: Chado feature tables, version 1.4
  tier: vendored
  kind: specification
  path: corpus/specifications/chado_1.4_feature_tables.sql
  format: PostgreSQL DDL
  source: GMOD Chado
  license: Artistic-2.0
  bytes: 5336
  md5: b39f9c0efeef2475c87cc9776f4e6726
  origin_url: https://github.com/GMOD/Chado/blob/1.4/schemas/1.4/default_schema.sql
  selection: 'the 7 feature tables extracted from a 2.2 MB schema: feature, featureloc, featureprop, feature_dbxref,
    feature_relationship, feature_cvterm, feature_synonym'
  retrieved: '2026-09-11'
  notes: The relational ancestor, and the fourth artifact in the link list that prompted this corpus.
    Chado keeps the feature separate from its location and expresses parent and child through a relationship
    table. Neither is nested, which is the same normalization a flat-table profile requires.
- id: spec-kbase-cdm-feature
  label: KBase Common Data Model, bioentity module with the Feature class
  tier: vendored
  kind: specification
  path: corpus/specifications/kbase_cdm_bioentity.yaml
  format: LinkML
  source: KBase
  license: MIT
  bytes: 22093
  md5: 2e30ced8ef62e424e8b06bb87d25c5a4
  origin_url: https://github.com/kbase/cdm-schema/blob/main/src/linkml/cdm_bioentity.yaml
  selection: the module holding Feature, EncodedFeature, Contig, ContigCollection, Protein and CdsPhaseType
  retrieved: '2026-09-11'
  notes: Vendorable because MIT. Constrains feature type to Sequence Ontology accessions under sequence_feature
    and carries provenance slots for source database and protocol, which the other models do not. No sample
    feature tables were found in that repository, so this is the model without data.
- id: emsl-monet
  label: EMSL MONet soil molecular data
  tier: linked
  format: various, not primarily feature tables
  source: EMSL at PNNL
  producer: MONet 1000 Soils pilot and later proposal calls
  producer_kind: observation network
  origin_url: https://www.emsl.pnnl.gov/monet
  data_portal_url: https://sc-data.emsl.pnnl.gov/
  jgi_portal_url: https://genome.jgi.doe.gov/portal/MONSoiUsFacility/MONSoiUsFacility.info.html
  paper_url: https://www.biorxiv.org/content/10.1101/2022.12.12.520098
  license: EMSL data policy; published without embargo
  retrieved: '2026-09-11'
  notes: Published on EMSL Science Central without embargo. Sequence-derived products route through JGI
    portals, which need an account. Recorded because BASALT, the LinkML model below, covers MONet.
- id: emsl-basalt
  label: EMSL BASALT schema
  tier: not located
  format: LinkML, reported
  source: EMSL at PNNL
  producer: EMSL
  producer_kind: schema
  license: unknown
  retrieved: '2026-09-11'
  notes: 'Reported as EMSL''s LinkML for MONet and parts of the analysis database, with a flattened version
    coming from James Carr. Searched GitHub across the ber-data and microbiomedata organizations and the
    open web on 2026-09-11 and found nothing public. Every code hit for ''basalt'' in those organizations
    is the rock, in environmental value sets. This is an open gap, not a dead end: ask EMSL directly.'
- id: tool-gffread
  label: gffread, GFF3 to GTF conversion
  tier: linked
  kind: tool
  format: GFF3 and GTF
  source: Cufflinks and StringTie lineage
  producer: Geo Pertea
  producer_kind: converter
  origin_url: https://github.com/gpertea/gffread
  license: MIT
  retrieved: '2026-09-11'
  notes: The conversion most people actually run between the two column 9 grammars. A parser and rewriter,
    not a data model.
- id: tool-gff3sort
  label: GFF3sort, sorting for tabix indexing
  tier: linked
  kind: tool
  format: GFF3
  source: billzt
  producer: billzt
  producer_kind: sorter
  origin_url: https://github.com/billzt/gff3sort
  license: not stated
  retrieved: '2026-09-11'
  notes: Exists because feature order is underspecified and browsers need a particular one. A parser and
    rewriter, not a data model.
- id: tool-aegis
  label: AEGIS, annotation extraction and genomic integration
  tier: linked
  kind: tool
  format: GFF and GTF
  source: preprint, 2025-12
  producer: AEGIS authors
  producer_kind: parser and normalizer
  origin_url: https://www.biorxiv.org/content/10.64898/2025.12.04.692274v1.full
  license: not stated
  retrieved: '2026-09-11'
  notes: Newest of the normalizers. Claims it parses files that make the other tools fail. Recorded as
    evidence that the problem is still open in 2025, not as a recommendation; not evaluated here.
- id: derived-cds-phase-illegal
  label: 'Derived malformed GFF3: phase in column 8 of the first CDS set to 3, outside the permitted 0,
    1 and 2'
  tier: derived
  path: corpus/fixtures/malformed/cds_phase_illegal.gff3
  format: GFF3 (invalid)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6888
  md5: 8a878e6455e02ca02069f8562c458b5f
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: phase in column 8 of the first CDS set to 3, outside the permitted 0, 1 and 2
  validity: 'invalid: GFF3 permits only 0, 1 or 2 in the phase column'
  why_it_matters: A validator catches this one. Paired with cds_phase_biologically_wrong, which no syntax-only
    validator catches, the two show that the AgBioData phase failure is not a syntax problem.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-dangling-parent
  label: 'Derived malformed GFF3: Parent of the first CDS changed to an ID defined nowhere in the file'
  tier: derived
  path: corpus/fixtures/malformed/dangling_parent.gff3
  format: GFF3 (invalid)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6889
  md5: a2803ead53e8e2058e6e208302a2295e
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: Parent of the first CDS changed to an ID defined nowhere in the file
  validity: 'invalid: GFF3 requires every Parent to resolve within the file'
  why_it_matters: Breaks the parent and child graph without breaking the line syntax, so it parses cleanly.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-duplicate-id
  label: 'Derived malformed GFF3: the first sequence_alteration given the first gene''s ID, gene-phiX174p04,
    so one ID names two features of different type'
  tier: derived
  path: corpus/fixtures/malformed/duplicate_id.gff3
  format: GFF3 (invalid)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6975
  md5: a6a3044a26dc1715604b2d701f3c3492
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: the first sequence_alteration given the first gene's ID, gene-phiX174p04, so one ID names
    two features of different type
  validity: 'invalid: a repeated ID is only legal across lines describing one feature, and these differ
    in type'
  why_it_matters: A repeated ID is legal when every line describes one discontinuous feature, so a duplicated
    row would NOT be a violation. Colliding the ID across two different feature types is, which is the
    distinction this fixture exists to hold.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-no-version-pragma
  label: 'Derived malformed GFF3: the ##gff-version 3 line removed; every other header line preserved'
  tier: derived
  path: corpus/fixtures/malformed/no_version_pragma.gff3
  format: GFF3 (invalid)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6860
  md5: 38c97705f13ccdb4f2b486b121c4649f
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: 'the ##gff-version 3 line removed; every other header line preserved'
  validity: 'invalid: GFF3 requires the version directive on line 1'
  why_it_matters: Without the directive a parser cannot tell GFF3 from GFF2, and the column 9 grammar
    differs between them.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    This is the documented exception among the derived files: its one change removes a header directive
    rather than altering a data row, so its header is deliberately NOT byte-identical to the source. Every
    other header line is preserved, and provenance comments are appended after the source''s own ### terminator.'
- id: derived-start-after-end
  label: 'Derived malformed GFF3: columns 4 and 5 of the first gene swapped, so start is greater than
    end'
  tier: derived
  path: corpus/fixtures/malformed/start_after_end.gff3
  format: GFF3 (invalid)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6897
  md5: 64d3964fb27e77e73ebaaacadff2ad93
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: columns 4 and 5 of the first gene swapped, so start is greater than end
  validity: 'invalid: GFF3 requires start <= end on every feature, circular included'
  why_it_matters: 'GFF3 requires start no greater than end on every feature. A circular genome does not
    change that: wraparound is expressed by extending end past the landmark length. phiX174 carries Is_circular=true,
    which is why the rule is worth stating rather than assuming.'
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-unescaped-semicolon
  label: 'Derived malformed GFF3: a raw semicolon inside a Note value on the first CDS, where GFF3 requires
    %3B'
  tier: derived
  path: corpus/fixtures/malformed/unescaped_semicolon.gff3
  format: GFF3 (invalid)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6977
  md5: 2f925867ea9489da98d712f4453cf511
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: a raw semicolon inside a Note value on the first CDS, where GFF3 requires %3B
  validity: 'invalid: the text after the separator is a bare segment with no tag=value form, which a conforming
    parser must reject'
  why_it_matters: The loud version. The text after the unescaped separator has no tag=value form, so a
    conforming parser must reject the record. Paired with unescaped_semicolon_silent, which a parser accepts.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-cds-phase-biologically-wrong
  label: 'Derived edge case: phase in column 8 of the first CDS changed from 0 to another permitted value'
  tier: derived
  path: corpus/fixtures/edge-cases/cds_phase_biologically_wrong.gff3
  format: GFF3 (valid; wrong or refused in practice)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 7017
  md5: 45275cca6480591ccb7b64ac2b41a6cb
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: phase in column 8 of the first CDS changed from 0 to another permitted value
  validity: 'VALID GFF3: 0, 1 and 2 are all legal, so no syntax-only validator catches this; translating
    against the reference protein does. The annotation is biologically wrong rather than malformed'
  why_it_matters: 'The failure the AgBioData group leads with, and no syntax-only validator can catch
    it: 0, 1 and 2 are all legal phases, so identical coordinates with a different phase translate to
    a different protein while the file stays valid. Translating against the reference protein does catch
    it.'
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-multiple-parents
  label: 'Derived edge case: a second Parent, gene-phiX174p06, added beside rna-phiX174p06; both are defined
    in this file'
  tier: derived
  path: corpus/fixtures/edge-cases/multiple_parents.gff3
  format: GFF3 (valid; wrong or refused in practice)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 6940
  md5: 998386f8cac265ef8f21a7006802b2e8
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: a second Parent, gene-phiX174p06, added beside rna-phiX174p06; both are defined in this file
  validity: 'VALID GFF3: multiple parents are permitted, and many tools refuse them anyway'
  why_it_matters: Legal GFF3 that many tools refuse. The AgBioData recommendations explicitly ask that
    parsers keep supporting it.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: derived-unescaped-semicolon-silent
  label: 'Derived edge case: an unescaped semicolon in a Note value whose tail is itself a valid lowercase
    tag=value pair'
  tier: derived
  path: corpus/fixtures/edge-cases/unescaped_semicolon_silent.gff3
  format: GFF3 (valid; wrong or refused in practice)
  source: derived in this repository
  producer: scripts/make_malformed.py
  producer_kind: derived test case
  bytes: 7030
  md5: a2edcb385cca680c823427cf8f54f894
  derived_from: corpus/sources/ncbi-refseq/ncbi_refseq_phix174_GCF_000819615.1.gff
  mutation: an unescaped semicolon in a Note value whose tail is itself a valid lowercase tag=value pair
  validity: 'VALID GFF3: the Note is silently truncated and a second attribute appears that nobody wrote;
    no validator reports anything because nothing is malformed'
  why_it_matters: The dangerous version of the escaping problem. The tail of the unescaped value is itself
    a valid lowercase tag=value pair, so the file stays valid, the intended Note is truncated, and an
    attribute nobody wrote now exists. Nothing reports anything, which is why this sits in edge-cases
    rather than malformed.
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/make_malformed.py
  retrieved: '2026-09-11'
  license: public domain (derived from a US Government work)
  notes: 'Derived, not observed in the wild, so it is not evidence about what any real producer emits.
    The file is its source preserved line for line with exactly one change to one data row, plus provenance
    comments, single-hash lines appended after the source''s own ### terminator, so the header region
    is byte-identical.'
- id: model-gff-schema
  label: GFF3 LinkML schema, biodatamodels
  tier: linked
  kind: model
  format: LinkML
  source: biodatamodels
  producer: biodatamodels (single contributor)
  producer_kind: data model
  origin_url: https://github.com/biodatamodels/gff-schema
  license: no license file in the repository
  why_not_vendored: no license file at all, so the schema cannot be copied here; compared instead in docs/model-comparison.md
  retrieved: '2026-09-11'
  notes: 'Self-described DRAFT. Last commit 2021-06-03, ten commits all from a single contributor, one
    open issue, number 1, from 2021-02-11. The only one of the four models that models a GFF3 file rather
    than a feature table: a gff document class holds features and sequence entries side by side. Also
    the only one modeling the header directives and pragmas as data, and the only one with an explicit
    seqid slot ranged over a model class. Its feature-class description is identical character for character
    to NMDC''s GenomeFeature, and that string is not in the GFF3 specification. First of the four artifacts
    in the 2026-09-08 link list that prompted this corpus.'
- id: spec-img-pipeline
  label: JGI IMG annotation pipeline documentation, version 5
  tier: linked
  kind: specification
  format: prose specification of GFF output
  source: JGI IMG
  producer: IMG annotation pipeline
  producer_kind: specification
  origin_url: https://img.jgi.doe.gov/docs/pipelineV5/
  license: not stated
  retrieved: '2026-09-11'
  notes: Specifies the GFF output of the pipeline that produced the original 14-file NMDC GFF sample, so it
    is the normative source for what those files are supposed to contain. Added to the reading list at
    the 2026-09-11 kickoff. Annotation specifics are behind a click-through from that page.
- id: tool-bcbb
  label: Blue Collar Bioinformatics, GFF parser
  tier: linked
  kind: tool
  format: GFF3
  source: chapmanb
  producer: Brad Chapman
  producer_kind: parser
  origin_url: https://github.com/chapmanb/bcbb
  license: see repository
  retrieved: '2026-09-11'
  notes: Named at the 2026-09-11 kickoff as an existing GFF parsing library. Recorded as an example of
    what consumers run, not as a recommendation; not evaluated here.
- id: biopython-blat-bed12
  label: BLAT v34 genomic alignments in BED12
  tier: vendored
  path: corpus/sources/biopython/blat_34_hg19.bed
  format: BED12
  source: Biopython pinned test corpus
  producer: BLAT v34 alignments distributed as BED12 by Biopython
  producer_kind: upstream example archive
  source_version: Biopython 1.85; query label hg19_dna; target assembly not declared in BED
  bytes: 1313
  md5: 314b31a371e6ab878b2a6285b597aa61
  sha256: c845389ca7e71cd9c3940275ef01730c6ed6805e8944a86898224aec909afc52
  origin_url: https://raw.githubusercontent.com/biopython/biopython/668de08f73fca7f8336049dfd78716d7dd095f21/Tests/Blat/psl_34_004.bed
  retrieved: '2026-09-21'
  license: Biopython License Agreement
  license_file: corpus/sources/biopython/LICENSE.rst
  license_url: https://github.com/biopython/biopython/blob/668de08f73fca7f8336049dfd78716d7dd095f21/LICENSE.rst
  vendored_form: Unmodified bytes from the pinned upstream file
  observation: Nineteen real-sequence alignment records with names, scores, both strands, thick spans
    and ordered blocks. The target assembly is not inferred from the query name.
- id: biopython-insdc-plant-records
  label: INSDC plant gene and transcript annotations in GenBank flat files
  tier: vendored
  path: corpus/sources/biopython/cor6_6.gb
  format: GenBank
  source: Biopython pinned test corpus
  producer: Original INSDC submitters identified in each record; retained by Biopython
  producer_kind: upstream example archive
  source_version: Biopython 1.85; X55053.1, X62281.1, M81224.1, AJ237582.1, L31939.1, AF297471.1
  bytes: 14967
  md5: 220a4da99e1150cd5e941856133a2627
  sha256: 01b4e193b71344752a96e2b37e886118a33c07b16c3684741c31eb3efeaf60b8
  origin_url: https://raw.githubusercontent.com/biopython/biopython/668de08f73fca7f8336049dfd78716d7dd095f21/Tests/GenBank/cor6_6.gb
  retrieved: '2026-09-21'
  license: Biopython License Agreement
  license_file: corpus/sources/biopython/LICENSE.rst
  license_url: https://github.com/biopython/biopython/blob/668de08f73fca7f8336049dfd78716d7dd095f21/LICENSE.rst
  vendored_form: Unmodified bytes from the pinned upstream file
  observation: Six deposited sequence records, including joined and partial locations and repeated qualifiers.
    This is GenBank Location/Qualifiers syntax, not an NCBI five-column submission table.
- id: nmdc-biosample-pfam
  label: Worked biosample pfam source
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-5xxrm214.2_pfam.gff
  format: GFF3
  source: NMDC
  producer: NMDC IMG annotation workflow nmdc:wfmgan-11-5xxrm214.2
  bytes: 102820
  md5: 00201b9a313e41413615b36fe3624fa9
  sha256: c143822c0ddad8bc6161b6a6a27a2c2c0fa806c986052557762e3a3aed6c14bf
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-2t8ft192/nmdc:wfmgan-11-5xxrm214.2/nmdc_wfmgan-11-5xxrm214.2_pfam.gff
  origin_record: nmdc:dobj-11-1afz5v98
  generated_by: nmdc:wfmgan-11-5xxrm214.2
  retrieved: '2026-09-21'
  license: CC BY 4.0
  selection: Companion evidence for the existing worked biosample; unchanged full artifacts
- id: nmdc-biosample-structural
  label: Worked biosample structural source
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-5xxrm214.2_structural_annotation.gff
  format: GFF3
  source: NMDC
  producer: NMDC IMG annotation workflow nmdc:wfmgan-11-5xxrm214.2
  bytes: 237883
  md5: 34456b0b386dacef212961f044640229
  sha256: 38340ec5f87db113c0c14a98fe58e1d1081a0d7f71d1a85891b59690cb7355ca
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-2t8ft192/nmdc:wfmgan-11-5xxrm214.2/nmdc_wfmgan-11-5xxrm214.2_structural_annotation.gff
  origin_record: nmdc:dobj-11-a1p22v52
  generated_by: nmdc:wfmgan-11-5xxrm214.2
  retrieved: '2026-09-21'
  license: CC BY 4.0
  selection: Companion evidence for the existing worked biosample; unchanged full artifacts
- id: nmdc-biosample-proteins
  label: Worked biosample proteins source
  tier: vendored
  path: corpus/sources/nmdc/nmdc_wfmgan-11-5xxrm214.2_proteins.faa
  format: FASTA (amino acids)
  source: NMDC
  producer: NMDC IMG annotation workflow nmdc:wfmgan-11-5xxrm214.2
  bytes: 245387
  md5: cd5df399a4a79f3efb46ab48c66b72af
  sha256: 49bd52721ccd054241b866a294cf3121bb04197cc0cd6f8f53fdd8b0ac5ed593
  md5_matches_source_record: true
  origin_url: https://data.microbiomedata.org/data/nmdc:omprc-11-2t8ft192/nmdc:wfmgan-11-5xxrm214.2/nmdc_wfmgan-11-5xxrm214.2_proteins.faa
  origin_record: nmdc:dobj-11-5z8x2253
  generated_by: nmdc:wfmgan-11-5xxrm214.2
  retrieved: '2026-09-21'
  license: CC BY 4.0
  selection: Companion evidence for the existing worked biosample; unchanged full artifacts
- id: "refseq-coelicolor-chromosome"
  label: "NCBI RefSeq NC_003888.3 chromosome annotations retrieved 2026-09-21"
  tier: "vendored"
  source: "NCBI RefSeq, Streptomyces coelicolor A3(2), NC_003888.3"
  format: "GFF3"
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  retrieved: "2026-09-21"
  path: "corpus/sources/ncbi-refseq/NC_003888.3_2026-09-21.gff3"
  bytes: 4691954
  md5: "d3e40d0d6ddd3f824b4a8b86a40131b3"
  sha256: "e49de7213e25d075e0b79541f8a49288a055ae87bdd355931d129c0573026858"
  origin_url: "https://www.ncbi.nlm.nih.gov/sviewer/viewer.fcgi?id=NC_003888.3&db=nuccore&report=gff3&retmode=text&from=5515000&to=5538000"
  notes: "Endpoint returned the complete chromosome annotation despite the from/to parameters. Retained unchanged; accession-version qualifies the sequence. Annotation snapshot is pinned by retrieval date and content hash, not accession version alone."
- id: "derived-actinorhodin-excerpt"
  label: "Actinorhodin SCO5071-SCO5092 gene/CDS excerpt with original coordinates"
  tier: "derived"
  source: "NCBI RefSeq, Streptomyces coelicolor A3(2), NC_003888.3"
  format: "GFF3"
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  retrieved: "2026-09-21"
  path: "corpus/derived-examples/actinorhodin.gff3"
  bytes: 13331
  md5: "ca9dbcdc33bdcba521d785934737c568"
  sha256: "2eaa1468d850ccfa147343148030182bbb95b0cd78d1e49d538c46697f3af14c"
  origin_url: "https://github.com/turbomam/feature-table-corpus/blob/main/scripts/bgc_example.py"
  derived_from: "refseq-coelicolor-chromosome"
  mutation: "Select genes with old_locus_tag SCO5071 through SCO5092 and their direct CDS children; retain original header, row bytes and chromosome coordinates; append selection provenance comments."
  validity: "valid GFF3: selected real rows; checked with GenomeTools 1.6.6"
  notes: "Biologically observed rows selected here; not a complete chromosome annotation. Both conversion round trips apply to this explicitly derived excerpt."
- id: "refseq-phix-genbank"
  label: "NCBI RefSeq phiX174 GenBank record with circular origin-crossing features"
  tier: "vendored"
  source: "NCBI RefSeq NC_001422.1, Escherichia phage phiX174"
  format: "GenBank nucleotide flat file"
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  retrieved: "2026-09-21"
  path: "corpus/sources/ncbi-refseq/NC_001422.1_2026-09-21.gb"
  bytes: 26238
  md5: "dad3a73118ab993fab0a836fb58fb593"
  sha256: "0a8099e207633f8c3ba998bfe2a198518e1de356c7956691101b2512324b807c"
  origin_url: "https://www.ncbi.nlm.nih.gov/sviewer/viewer.fcgi?id=NC_001422.1&db=nuccore&report=genbank&retmode=text"
  notes: "Unchanged whole 5386 bp circular nucleotide record, 32 features including six origin-crossing joins. Complements the independently retained GFF3 annotation; source hash pins annotation snapshot separately from sequence accession version."
- id: jgi-img-clostridium-cath-funfam
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) CATH functional families search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_cath_funfam.gff
  format: GFF3
  source: JGI Data Portal
  producer: HMMER 3.1b2 search of CATH functional families
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 1329740
  md5: 05c603dd179ca0c635338b3ffd919173
  sha256: 25bfc4fa96d4bc36d5e1b6c09c8c2e6ed29b856609989a6db72a90b9c383086b
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33d6263bf2148832ef7/Ga0423362_cath_funfam.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33d6263bf2148832ef7
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-cleavage-sites
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) SignalP cleavage sites
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_cleavage_sites.gff
  format: GFF3
  source: JGI Data Portal
  producer: SignalP 4.1 signal peptide cleavage site prediction
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 18175
  md5: 168c2bdc36962433f1cd43f48e1ddc5e
  sha256: 142c39926b279ac7e96a20d44d27cd872b9ded71f2395e6de34f6d9265be54d3
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33b6263bf2148832ed9/Ga0423362_cleavage_sites.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33b6263bf2148832ed9
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-cog
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) COG search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_cog.gff
  format: GFF3
  source: JGI Data Portal
  producer: COG database search
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 927609
  md5: a85084105705f69d5d01b6747ccd2659
  sha256: cf2d624dfb7d691d9825ac2b1ef0ca29bf62243dd45c3b1c613b3137250609bf
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33a6263bf2148832ecc/Ga0423362_cog.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33a6263bf2148832ecc
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-crt
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) CRT CRISPR arrays
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_crt.gff
  format: GFF3
  source: JGI Data Portal
  producer: CRISPR Recognition Tool 1.8.2
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 690
  md5: 06c8ff10e31d69d6394aa6c0bf788d66
  sha256: a3d42be8d21e9e32e2a6edcee9382a92d24c989804583f3df4b97ee7feebcddc
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832eef/Ga0423362_crt.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832eef
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-functional-annotation
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) functional roll-up
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_functional_annotation.gff
  format: GFF3
  source: JGI Data Portal
  producer: JGI IMG annotation pipeline, functional roll-up
  producer_kind: roll-up
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 1386145
  md5: dc12d0593b5f3c52bf173331ad89052e
  sha256: ffbcca39c15830728367ea8fba80252d239842394fdd05bd10bf4d2fd834632a
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832eeb/Ga0423362_functional_annotation.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832eeb
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-genemark
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) GeneMark gene calls
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_genemark.gff
  format: 'GFF3 (declares ##gff-version 2)'
  source: JGI Data Portal
  producer: GeneMark.hmm-2 v1.05, gene caller
  producer_kind: gene caller
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 481892
  md5: 2b2daa60b138d433946cbca97f8e224e
  sha256: d4653a8edbd6d2c68922a842c5269dd558701cf6d85e626e8faef4648c85338a
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832eed/Ga0423362_genemark.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832eed
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: 'First line is ##gff-version 2 while column 9 uses GFF3 tag=value attributes.'
- id: jgi-img-clostridium-ko-ec
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) KEGG Orthology and EC assignment
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_ko_ec.gff
  format: GFF3
  source: JGI Data Portal
  producer: lastal 983 search for KEGG Orthology and Enzyme Commission numbers
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 559474
  md5: 40206bf13bbb3a052e33150a626b96b8
  sha256: 6949c6eda98e544d62556d67efa152d3219b57bcf9b966083ea27b7c445afbd2
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33b6263bf2148832eda/Ga0423362_ko_ec.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33b6263bf2148832eda
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-pfam
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) Pfam search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_pfam.gff
  format: GFF3
  source: JGI Data Portal
  producer: HMMER 3.1b2 search of Pfam
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 1045944
  md5: f00a1a55fab731e96ba29a56f6ee0e70
  sha256: 1d9e3a23141ef750964b8b18591bf16bb41eeb035bb7a14b2aa6f7a7a06dc151
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33b6263bf2148832ede/Ga0423362_pfam.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33b6263bf2148832ede
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-prodigal
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) Prodigal gene calls
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_prodigal.gff
  format: GFF3
  source: JGI Data Portal
  producer: Prodigal v2.6.3, gene caller
  producer_kind: gene caller
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 452840
  md5: b5fa3db7bb855bad9d11d1d1bdb1c16a
  sha256: 1b5352af9a8722e743c07fe152c43ffddd01dd4e7fc468785f21f4526dfc25fc
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33a6263bf2148832ec7/Ga0423362_prodigal.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33a6263bf2148832ec7
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-rfam
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) Rfam search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_rfam.gff
  format: GFF3
  source: JGI Data Portal
  producer: INFERNAL 1.1.2 search of Rfam
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 21904
  md5: df566e904bb9f63f626e35833305c066
  sha256: 2fc7df38b341339352a3252c3c118d0da544be32cface6418f3c7f57a87bda0b
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832eea/Ga0423362_rfam.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832eea
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-rfam-misc-bind-misc-feature-regulatory
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) Rfam misc_bind, misc_feature and
    regulatory subset
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_rfam_misc_bind_misc_feature_regulatory.gff
  format: GFF3
  source: JGI Data Portal
  producer: INFERNAL 1.1.2 search of Rfam
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 14267
  md5: fe8c470ac9d213e775903248da4fa629
  sha256: 7e62e1d79ddb04641bd6a8e4ce5598322e65579f13abe370b8c68efdfca1fa1a
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33a6263bf2148832ec9/Ga0423362_rfam_misc_bind_misc_feature_regulatory.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33a6263bf2148832ec9
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-rfam-ncrna-tmrna
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) Rfam ncRNA and tmRNA subset
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_rfam_ncrna_tmrna.gff
  format: GFF3
  source: JGI Data Portal
  producer: INFERNAL 1.1.2 search of Rfam
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 1318
  md5: fcc88bfab85559a26582c2fd1c781073
  sha256: 0f5c2ef75e78aed9f40c728189840ea5170ce39d431ec270a92d848340985446
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832ef0/Ga0423362_rfam_ncrna_tmrna.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832ef0
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-rfam-rrna
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) Rfam rRNA subset
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_rfam_rrna.gff
  format: GFF3
  source: JGI Data Portal
  producer: INFERNAL 1.1.2 search of Rfam
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 6319
  md5: 1dd5582e20a3cdc2a21d47a87022af9a
  sha256: 228990f4a7f67dcd009083f4d9a8279a9a5579ad9b9dfcbfd6c21e20b04220e3
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832ef1/Ga0423362_rfam_rrna.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832ef1
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-smart
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) SMART search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_smart.gff
  format: GFF3
  source: JGI Data Portal
  producer: SMART database search
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 432216
  md5: 5822395083abf3ab081144fd619e62af
  sha256: f927403d5d60401e3e596047729f2ddaf438710c82a28d6162af346ff05238ad
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832ee6/Ga0423362_smart.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832ee6
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-structural-annotation
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) structural roll-up
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_structural_annotation.gff
  format: GFF3
  source: JGI Data Portal
  producer: JGI IMG annotation pipeline, structural roll-up
  producer_kind: roll-up
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 586379
  md5: 4dc2b0d1f231720089b7f65381cb9527
  sha256: b05e90d843235abad80a4e491f652e5d7f4cf384b21fe89921cbbce21fcf5157
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33b6263bf2148832edc/Ga0423362_structural_annotation.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33b6263bf2148832edc
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-supfam
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) SUPERFAMILY search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_supfam.gff
  format: GFF3
  source: JGI Data Portal
  producer: HMMER 3.1b2 search of SUPERFAMILY
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 1381060
  md5: 05e8683a592b8d01e55d2163b957940f
  sha256: 95f7b5fdf782386f240e831b06eaf8cd1790d5b55ea98c537b0a10d1a21d95c6
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33b6263bf2148832edd/Ga0423362_supfam.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33b6263bf2148832edd
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-tigrfam
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) TIGRFAM search
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_tigrfam.gff
  format: GFF3
  source: JGI Data Portal
  producer: TIGRFAM database search
  producer_kind: database search
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 213597
  md5: 33bcbb51d5bbfd95fa05696431c153ee
  sha256: e3da798f08b67970d0b667d3c53e9518d190443576de3aad7805293cd2a2d387
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33c6263bf2148832ee7/Ga0423362_tigrfam.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33c6263bf2148832ee7
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-tmh
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) TMHMM transmembrane topology
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_tmh.gff
  format: GFF3
  source: JGI Data Portal
  producer: decodeanhmm 1.1g (TMHMM) transmembrane helix prediction
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 1000930
  md5: 73416d0bef09ac9ec8f522e5c7b1151b
  sha256: 4a848198f45e3244549c93dccabec89a99f5a9a73c7d2e7677f08159f52f1c1c
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33a6263bf2148832ec4/Ga0423362_tmh.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33a6263bf2148832ec4
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-clostridium-trna
  label: Clostridium acetobutylicum DJ311 IMG_AP-1268149 (Ga0423362) tRNAscan-SE tRNAs
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1268149/Ga0423362_trna.gff
  format: GFF3
  source: JGI Data Portal
  producer: tRNAscan-SE 2.0.5, feature finder
  producer_kind: feature finder
  source_version: IMG annotation pipeline, per-method GFF set with structural and functional roll-ups
  bytes: 13467
  md5: 44804feb3711da8923f4b146dcd25a33
  sha256: 340969f36a687dfb657553a70e471a757cfaf841ff3e33f4f300d5de0e2b46af
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/5ee7d33a6263bf2148832ecb/Ga0423362_trna.gff
  origin_record: IMG_AP-1268149
  origin_file_id: 5ee7d33a6263bf2148832ecb
  retrieved: '2026-09-25'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488021, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488021
  proposal: BrcProposal 2804
  proposal_acceptance_date: '2015-12-10'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
- id: jgi-img-bacillus-assembled-gff
  label: Bacillus sp. V-88 IMG_AP-1121004 IMG pipeline 4.14.0 assembled annotation GFF
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/106476.assembled.gff
  format: GFF3
  source: JGI Data Portal
  producer: JGI IMG annotation pipeline 4.14.0 (Prodigal, INFERNAL, HMMER)
  producer_kind: roll-up
  source_version: IMG pipeline 4.14.0, per 106476.pipeline_version.info in the same portal record
  bytes: 667730
  md5: 21b6400d7e9c82c01ea1a286baf01bf9
  sha256: e97219b1d1f70724b0b95c3652e4342c36a096c7ee1e2f6af68a92f3552fea4d
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/585823e07ded5e78cff8edcb/106476.assembled.gff
  origin_record: IMG_AP-1121004
  origin_file_id: 585823e07ded5e78cff8edcb
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: 'Column 7 holds 1 and -1 rather than + and -, and there is no ##gff-version line. Same assembly
    (Ga0139071) as 2708743150.gff, from an earlier pipeline generation.'
- id: jgi-img-bacillus-taxon-gff
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 taxon GFF
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.gff
  format: GFF3
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 taxon export
  producer_kind: roll-up
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 615004
  md5: f42ce397a9106100ca58d94ab4b93c76
  sha256: 79b7b72c28a46f63aabf897b1813836121e60a3a909ed80d932059983b0a093c
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac94/2708743150.gff
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac94
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Attributes are only ID, locus_tag and product; functional annotation is in the .tab.txt side
    tables, joined on gene_oid = ID.
- id: jgi-img-bacillus-cog-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 COG side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.cog.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 COG table
  producer_kind: database search
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 311598
  md5: 7ba882ac0614c94717387d9e27a09161
  sha256: a40ffd17341fb341ab14dc6ff6e6f116f6b2764b56ba824c440aaac30f18933e
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac92/2708743150.cog.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac92
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-img-bacillus-ipr-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 InterPro with GO side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.ipr.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 InterPro with GO table
  producer_kind: database search
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 686982
  md5: 1ef979df60827b86078f1b28cbe5be09
  sha256: d6e21482005c82ae047e0c7231ed54c8f1021acd31d7956538f9c9a97e34a2ef
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac86/2708743150.ipr.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac86
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-img-bacillus-ko-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 KEGG Orthology with EC side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.ko.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 KEGG Orthology with EC table
  producer_kind: database search
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 268565
  md5: a652e06193f50ee804291602ba4ff93c
  sha256: 460af986ffebe30975e3116640d15b8337072b97b7ac91163fdf5437fadd1f02
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac98/2708743150.ko.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac98
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-img-bacillus-pfam-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 Pfam side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.pfam.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 Pfam table
  producer_kind: database search
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 353141
  md5: 3ff86b2d280581152bed91d597863e60
  sha256: 0e8f50f96629e49c0214d3409a614566381c4e6a7c340596efb941a0219480b4
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac90/2708743150.pfam.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac90
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-img-bacillus-signalp-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 SignalP side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.signalp.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 SignalP table
  producer_kind: feature finder
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 5695
  md5: ea25b2aec9744ede3ec17189da9160e4
  sha256: f1acb5ca41fb4c95be046d3ac2ccaa1ab71361f0ff55e544df06f2fc7a723fef
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac8e/2708743150.signalp.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac8e
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-img-bacillus-tigrfam-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 TIGRFAM side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.tigrfam.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 TIGRFAM table
  producer_kind: database search
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 128243
  md5: a67c8b39aa9c601233531b5dcda46c1d
  sha256: c2d5123a3b0c84a479aa60787196dd0ba75d7d817057d289003981a108322067
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac84/2708743150.tigrfam.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac84
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-img-bacillus-tmhmm-table
  label: Bacillus sp. V-88 IMG_AP-1121004 img_core_v400 TMHMM side table
  tier: vendored
  path: corpus/sources/jgi-img/IMG_AP-1121004/2708743150.tmhmm.tab.txt
  format: TSV
  source: JGI Data Portal
  producer: JGI IMG, img_core_v400 TMHMM table
  producer_kind: feature finder
  source_version: img_core_v400 taxon bundle, taxon 2708743150
  bytes: 481653
  md5: 7b021eea69be390c4ab8eb0868c98a88
  sha256: bd6e40d7288d04141e82d78498c187836e590ac4edee73d96f8629e04323d333
  md5_matches_source_record: true
  origin_url: https://files.jgi.doe.gov/filedownload/6822952a9a4399cb6ee0ac88/2708743150.tmhmm.tab.txt
  origin_record: IMG_AP-1121004
  origin_file_id: 6822952a9a4399cb6ee0ac88
  retrieved: '2026-09-24'
  license: JGI Legacy Data Policy (proposal accepted before FY22); data_utilization_status Unrestricted;
    attribution requested by this repository
  license_url: https://jgi.doe.gov/sites/default/files/2025-01/Data%26Support_LegacyDataPolicy.pdf
  attribution: 'Cite the dataset, https://doi.org/10.25585/1488085, and include the JGI acknowledgment:
    These sequence data were produced by the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/
    in collaboration with the user community.'
  dataset_doi: 10.25585/1488085
  proposal: Csp2014Proposal 1312
  proposal_acceptance_date: '2013-09-20'
  origin_requires_login: true
  selection: Part of an organism-centric annotation set added for https://github.com/turbomam/feature-table-corpus/issues/53;
    unchanged full file
  barrier: Downloading from origin_url needs a JGI Data Portal login; the anonymous search API reports
    the md5
  notes: Not GFF. Tab-separated with a header row; rows are keyed by gene_oid, the ID of a row in 2708743150.gff.
- id: jgi-phytozome-populus-v4.1-gene-exons
  label: Populus trichocarpa v4.1 (Phytozome-533) gene_exons GFF3
  tier: restricted
  format: GFF3
  source: JGI Phytozome, Populus trichocarpa v4.1 (Phytozome-533)
  producer: JGI plant annotation pipeline (phytozomev13)
  producer_kind: reference annotation
  origin_url: https://files.jgi.doe.gov/search/?q=Phytozome-533
  jgi_file_id: 5d94dc9fc0d65a87debccfce
  file_name: Ptrichocarpa_533_v4.1.gene_exons.gff3.gz
  bytes: 9769649
  md5: 3fdcfdd5af213c5f4dde0c01f2cdbfda
  license: Public; cite Tuskan et al. 2006 (Science 313:1596) and Goodstein et al. 2012 (Nucleic Acids Res 40:D1178), and give JGI's acknowledgment as worded in corpus/derived-examples/README.md (which JGI data policy governs this genome is undetermined)
  barrier: requires a JGI account; 112.8 MB uncompressed (849,981 lines), too large to vendor
  notes: The source of the derived excerpt populus-v4.1-chr01-200-genes. md5 from the JGI files API, matching the downloaded file (2026-09-30).
  retrieved: '2026-09-29'
- id: populus-v4.1-chr01-200-genes
  label: Populus trichocarpa v4.1 first 200 Chr01 genes, gene_exons excerpt
  tier: derived
  source: JGI Phytozome, Populus trichocarpa v4.1 (Phytozome-533)
  format: GFF3
  license: Public; cite Tuskan et al. 2006 (Science 313:1596) and Goodstein et al. 2012 (Nucleic Acids Res 40:D1178), and give JGI's acknowledgment as worded in corpus/derived-examples/README.md (which JGI data policy governs this genome is undetermined)
  retrieved: '2026-09-29'
  path: corpus/derived-examples/populus-v4.1-chr01-200-genes.gene_exons.gff3
  bytes: 659259
  md5: 1832c9478a744a3e776cde9d1c154254
  sha256: 9cea4c1a1dcc4ef89186341513ae91035efb52b94d6d10e236fbf85fd849f679
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/populus_excerpt.py
  redistribution_basis:
    terms:
    - https://phytozome-next.jgi.doe.gov/api/db/properties/proteome/533
    read: '2026-10-01'
    allows: Phytozome's public record for the genome gives "data_restriction_policy" as "unrestricted" (read 2026-09-29 and again 2026-10-01). Which JGI data policy governs it is undetermined, because the JGI search API gives no proposal acceptance date (model/examples/jgi-inputs.yaml). The genome's own Ptrichocarpa_533_v4.1.DataReleasePolicy.html (JGI file 5d94dc9ec0d65a87debccfc7, behind a JGI login) asks only for citation, as docs/restricted-sources.md records. Mark accepted redistributing excerpts with those citations on 2026-10-01.
  derived_from: jgi-phytozome-populus-v4.1-gene-exons
  mutation: Select the first 200 genes on Chr01 in file order with all their mRNA, exon, CDS and UTR rows; retain the three directives, row bytes and chromosome coordinates; add the four provenance comment lines after the directives, the last giving the citations in corpus/derived-examples/README.md.
  validity: 'valid GFF3: selected source rows, not a complete genome annotation; checked with GenomeTools 1.6.6, and valid under the phytozome-gene-exons-gff3 dialect'
  notes: The real Phytozome input for the gene_exons dialect's clean round trip (https://github.com/turbomam/feature-table-corpus/issues/107). The full file is behind a JGI login and too large to vendor.
- id: jgi-phytozome-populus-v4.1-annotation-info
  label: Populus trichocarpa v4.1 (Phytozome-533) annotation_info table
  tier: restricted
  format: TSV
  source: JGI Phytozome, Populus trichocarpa v4.1 (Phytozome-533)
  producer: JGI plant annotation pipeline
  producer_kind: reference annotation
  origin_url: https://files.jgi.doe.gov/search/?q=Phytozome-533&file=annotation_info
  jgi_file_id: 5d94dc9ec0d65a87debccfcc
  file_name: Ptrichocarpa_533_v4.1.annotation_info.txt
  bytes: 8742751
  md5: aaf04aa3515d96748d1229d36ac7dfe4
  license: Public; cite Tuskan et al. 2006 (Science 313:1596) and Goodstein et al. 2012 (Nucleic Acids Res 40:D1178), and give JGI's acknowledgment as worded in corpus/derived-examples/README.md (which JGI data policy governs this genome is undetermined)
  barrier: requires a JGI account
  notes: The source of the derived excerpt populus-v4.1-chr01-200-genes-annotation-info. Its header swaps the KOG and ec labels; the JGI files API lists a sibling named ...annotation_info.txtwrong_header. md5 from the JGI files API, matching the downloaded file (2026-09-30).
  retrieved: '2026-09-29'
- id: populus-v4.1-chr01-200-genes-annotation-info
  label: Populus trichocarpa v4.1 first 200 Chr01 genes, annotation_info excerpt
  tier: derived
  source: JGI Phytozome, Populus trichocarpa v4.1 (Phytozome-533)
  format: TSV
  license: Public; cite Tuskan et al. 2006 (Science 313:1596) and Goodstein et al. 2012 (Nucleic Acids Res 40:D1178), and give JGI's acknowledgment as worded in corpus/derived-examples/README.md (which JGI data policy governs this genome is undetermined)
  retrieved: '2026-09-29'
  path: corpus/derived-examples/populus-v4.1-chr01-200-genes.annotation_info.txt
  bytes: 49295
  md5: 9d568c6fee654f4df3ee15b64de0dff7
  sha256: 229a9a60d3671cd37c2a566ed55520439167cfb5725f211b47b8a0f5b8bcd16a
  origin_url: https://github.com/turbomam/feature-table-corpus/blob/main/scripts/populus_excerpt.py
  redistribution_basis:
    terms:
    - https://phytozome-next.jgi.doe.gov/api/db/properties/proteome/533
    read: '2026-10-01'
    allows: Phytozome's public record for the genome gives "data_restriction_policy" as "unrestricted" (read 2026-09-29 and again 2026-10-01). Which JGI data policy governs it is undetermined, because the JGI search API gives no proposal acceptance date (model/examples/jgi-inputs.yaml). The genome's own Ptrichocarpa_533_v4.1.DataReleasePolicy.html (JGI file 5d94dc9ec0d65a87debccfc7, behind a JGI login) asks only for citation, as docs/restricted-sources.md records. Mark accepted redistributing excerpts with those citations on 2026-10-01.
  derived_from: jgi-phytozome-populus-v4.1-annotation-info
  mutation: Select the rows of the 200 genes in populus-v4.1-chr01-200-genes, in table order; retain the header and row bytes; add the four provenance comment lines after the header, the last giving the citations in corpus/derived-examples/README.md.
  validity: selected source rows, not a complete genome annotation; valid under the phytozome-annotation-info dialect
  notes: The real Phytozome input for the annotation_info dialect's clean round trip, with the gene_exons excerpt (https://github.com/turbomam/feature-table-corpus/issues/144). 293 rows, 2 of them with a peptideName naming another locus's transcript.
- id: smc-coelicolor-genome-secmet-gff
  label: JGI SMC genome GFF3 for Streptomyces coelicolor A3(2), RefSeq-style annotation with antiSMASH features merged in
  tier: linked
  format: GFF3
  source: JGI Secondary Metabolite Collaboratory (SMC), source smc_id 212512, GCA_000203835.1
  producer: SMC ingest of an NCBI annotwriter GFF3 plus antiSMASH v7.0.0beta1 and EMERALD v0.2.4.1 predictions (run 2023-06-08)
  producer_kind: database export
  origin_url: https://smc.jgi.doe.gov/v2/download/1888169
  origin_record: https://smc.jgi.doe.gov/v2/smc/212512
  license: not stated; SMC asks users to cite Udwary et al., Nucleic Acids Res, https://doi.org/10.1093/nar/gkae1060
  why_not_vendored: no license or terms of use stated, and the annotations are SMC's own work; linked until SMC (Dan Udwary or Drew Doering) agrees to copies. Also 19,285,049 bytes.
  notes: File GCA_000203835.1.filtered.secmet.gff, 19,285,049 bytes, md5 8e69d11eb9c8db64746b86e71f77ee38, fetched without a login. Contig names are GenBank (AL645882.2), not RefSeq (NC_003888.3). The same chromosome as refseq-coelicolor-chromosome, so SMC and RefSeq can be compared feature by feature (https://github.com/turbomam/feature-table-corpus/issues/166).
  retrieved: '2026-10-09'
- id: smc-actinorhodin-bgc-annotations
  label: JGI SMC per-tool GFF3 annotations of the actinorhodin BGC (SMC bgc_id 2559962)
  tier: linked
  format: GFF3
  source: JGI Secondary Metabolite Collaboratory (SMC), bgc_id 2559962 on AL645882.2:5496474-5567376, chemical product "actinorhodin", class Polyketide/T2PKS
  producer: one GFF3 per annotation tool, served by the SMC API
  producer_kind: database export
  origin_url: https://smc.jgi.doe.gov/v2/bgc-tracks/2559962
  origin_record: https://smc.jgi.doe.gov/v2/bgc/2559962
  license: not stated; SMC asks users to cite Udwary et al., Nucleic Acids Res, https://doi.org/10.1093/nar/gkae1060
  why_not_vendored: no license or terms of use stated; small enough to vendor if SMC agrees
  notes: "Five files, each from https://smc.jgi.doe.gov/v2/bgc/annotation/<id>, fetched without a login: antiSMASH v7.0.0beta1 12203038 (48,373 bytes, 89 feature rows, md5 c8718e0aede2bb5378c9d2b697a9b871); InterProScan 12203044 (58,781 bytes, 250 rows, md5 d600afd067f3307512c37e0fdb83d054); CDDSearch 12203050 (1,355 bytes, 9 rows, md5 30b80113f783b8c32a53eaa07912441f); EMERALD v0.2.4.1 12203043 (365 bytes, 2 rows incl. nearest MIBiG BGC0000194, md5 f188adcb8f443b47c9058ff713d7cee4); GenBank 12203052 (162 bytes, the source region row only, md5 2d68775ea8e2cd77671cb9cc8944d980). antiSMASH's proto_core ends at 5,532,424, the end of SCO5088 in derived-actinorhodin-excerpt."
  retrieved: '2026-10-09'
- id: refseq-viral-nc-067194-1
  label: "NCBI RefSeq GFF3, Carjivirus communis (metagenome-assembled), complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_067194.1_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_067194.1
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 40225
  md5: 7d940187128aebda168a704471878a93
  sha256: 8f1fe80f6b6aaf28324602dcf917cfc1c22aac1bff7dbf56de1d05aec0e5934b
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_067194.1&retmode=text
  origin_record: NC_067194.1
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "circular DNA, 97,065 bp, assembled from a human fecal metagenome (`metagenomic=true`). Gene and CDS OIE19_gp01 end at 97,165, past the sequence length, the GFF3 way of writing an origin-crossing feature. CDS YP_010509440.1 is two rows with one ID. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-001497-2
  label: "NCBI RefSeq GFF3, Cauliflower mosaic virus, complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_001497.2_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_001497.2
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 3816
  md5: 6fb5f311c02c6b21b217a65c189cb40a
  sha256: 501da196ff1a4f0270ad9191cce10743fbc785335574cf881627805c711ff473
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_001497.2&retmode=text
  origin_record: NC_001497.2
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "circular DNA, 8,024 bp. CDS NP_597788.1 is partial at its start (`partial=true;start_range=.,2`). Two sequence_feature rows mark single-stranded gaps, and one modified_DNA_base row. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-021798-1
  label: "NCBI RefSeq GFF3, Cellulophaga phage phi17:2, complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_021798.1_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_021798.1
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 100889
  md5: d6181af99357e78dca3ab4d3a3775e43
  sha256: 118d793bc949b8bff464daaead26d6164b024bf6eeaa8dc16c622fdd08375d39
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_021798.1&retmode=text
  origin_record: NC_021798.1
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "circular DNA, 145,343 bp. 22 tRNA genes with exon children and one pseudogenic_tRNA. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-014649-1
  label: "NCBI RefSeq GFF3, Acanthamoeba polyphaga mimivirus, complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_014649.1_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_014649.1
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 1301423
  md5: bf738642c916bc905a73ebaf7152bfe6
  sha256: 312564505aa0ed80cd12e7fe4976104c38153fa8b18b91febfb065209483c8b8
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_014649.1&retmode=text
  origin_record: NC_014649.1
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "linear DNA, 1,181,549 bp. 1,094 promoter, 732 stem_loop and 1,086 UTR rows. Three spliced CDS span two to four rows each with one ID. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-001417-2
  label: "NCBI RefSeq GFF3, phage MS2 genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_001417.2_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_001417.2
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 2961
  md5: 60665e37a536d173ca3142bd30492bd7
  sha256: e617dc472015e0e180927e49cc615f85658eefdcea262df701ce64078e79a718
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_001417.2&retmode=text
  origin_record: NC_001417.2
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "linear single-stranded RNA genome (`mol_type=genomic RNA`), 3,569 bp. Overlapping CDS and two sequence_difference rows. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-003291-2
  label: "NCBI RefSeq GFF3, Bacteriophage PSA complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_003291.2_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_003291.2
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 28319
  md5: 459f6217809f136dcce9122fb5001cc7
  sha256: e56f8d85bd479e02ec49b55560ef6ec26fcefd89abc1d680368888a8ee248b41
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_003291.2&retmode=text
  origin_record: NC_003291.2
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "linear DNA, 37,618 bp. Two CDS with `Note=ribosomal slippage`, each written as two rows with one ID, the second row at phase 1. A sequence_feature marks the cohesive end. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-001421-2
  label: "NCBI RefSeq GFF3, Enterobacteria phage PRD1, complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_001421.2_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_001421.2
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 12911
  md5: 3b2cfc81e21222ecc3ff887822fb89c0
  sha256: 83819caff39bd86743a51ae33832d25d83f2b92be28089442c44fce434ad48de
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_001421.2&retmode=text
  origin_record: NC_001421.2
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "linear DNA, 14,927 bp, 31 genes and CDS, many overlapping. Byte-identical to the copy @simroux shared."
- id: refseq-viral-nc-001890-1
  label: "NCBI RefSeq GFF3, Enterobacteria phage Qbeta, complete genome"
  tier: vendored
  path: corpus/sources/ncbi-refseq/NC_001890.1_2026-10-30.gff3
  format: GFF3
  gff_spec_version: '1.21'
  source: NCBI RefSeq NC_001890.1
  producer: NCBI annotwriter
  producer_kind: reference annotation
  bytes: 3090
  md5: f0700d338b45f01a823d21422e19f499
  sha256: d2cebe06fdf3f5db1d2c35a12867ccc90dff0c9bea48564fda62f8037cf6623c
  origin_url: https://www.ncbi.nlm.nih.gov/sviewer/viewer.cgi?db=nuccore&report=gff3&id=NC_001890.1&retmode=text
  origin_record: NC_001890.1
  selection: one of nine NCBI virus GFF3 files chosen by @simroux to show unusual viral genome features (https://github.com/turbomam/feature-table-corpus/issues/179)
  retrieved: '2026-10-30'
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "linear single-stranded RNA genome, 4,215 bp. CDS NP_046750.1 is the readthrough product of the coat protein CDS and carries `transl_except=(pos:1742..1744%2Caa:Trp)`. Byte-identical to the copy @simroux shared."
- id: ensembl-scer-r64-mito-gff3
  label: "Ensembl GFF3, Saccharomyces cerevisiae R64-1-1 mitochondrial chromosome (release 63)"
  tier: "vendored"
  path: "corpus/sources/ensembl/Saccharomyces_cerevisiae.R64-1-1.63.chromosome.Mito.gff3"
  format: "GFF3"
  source: "Ensembl release 63"
  producer: "Ensembl GFF3 dump (annotation from SGD)"
  producer_kind: "reference annotation"
  bytes: 44006
  md5: "ae0b306fe79ed678fdf992bd369f5be1"
  sha256: "7f12e307e0ec895731b953e46638d6ef516e50fe5c714f36a0650a3a04f08b15"
  origin_url: "https://ftp.ensembl.org/pub/current_gff3/saccharomyces_cerevisiae/Saccharomyces_cerevisiae.R64-1-1.63.chromosome.Mito.gff3.gz"
  origin_record: "AJ011856.1"
  selection: "smallest Ensembl GFF3 found that writes ### between gene blocks, #! header lines, a circular sequence and CDS IDs spread over several rows; stored uncompressed"
  retrieved: "2026-11-06"
  license: "Ensembl places no restrictions on its data; see https://mart.ensembl.org/info/about/legal/disclaimer.html (annotation originates from SGD)"
  notes: "circular (Is_circular=true), 85,779 bp; 56 ### sync marks, 5 #! lines, 2 ## directives, 257 feature rows; CDS:Q0045 spans 8 rows. Same chromosome as the NCBI GenBank record NC_001224.1."
- id: refseq-scer-mito-nc-001224-1-genbank
  label: "NCBI RefSeq GenBank, Saccharomyces cerevisiae S288c mitochondrion (NC_001224.1)"
  tier: "vendored"
  path: "corpus/sources/ncbi-refseq/NC_001224.1_2026-11-06.gb"
  format: "GenBank"
  source: "NCBI RefSeq NC_001224.1"
  producer: "NCBI"
  producer_kind: "reference annotation"
  bytes: 176483
  md5: "4be624c7619b79bd7236becc171c3abd"
  sha256: "7c3ae57d2b3fe253b3f135c6ca3a9dbe69676a31f21135a17eb1866488e0bde0"
  origin_url: "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nuccore&id=NC_001224.1&rettype=gbwithparts&retmode=text"
  origin_record: "NC_001224.1"
  selection: "the GenBank record for the same chromosome as the Ensembl mitochondrial GFF3, for interconversion"
  retrieved: "2026-11-06"
  license: "Freely available NCBI data; see https://www.ncbi.nlm.nih.gov/home/about/policies/"
  notes: "circular, 85,779 bp, 19 CDS."
- id: bakta-mgenitalium-g37-gff3
  label: "Bakta GFF3, Mycoplasmoides genitalium G37 (NC_000908.2)"
  tier: "vendored"
  path: "corpus/sources/bakta/Mycoplasma_genitalium_G37_NC_000908.2_bakta-1.12.2.gff3"
  format: "GFF3"
  source: "Bakta v1.12.2 run on NCBI RefSeq NC_000908.2"
  producer: "Bakta 1.12.2"
  producer_kind: "annotation pipeline"
  bytes: 687506
  md5: "c8db6e5fb6a7094319100356fec35c50"
  sha256: "5128f627c23afed24319049a1e59c26bd1564c6c7af35fbc5af35eadcfa9242f"
  origin_url: "https://www.ncbi.nlm.nih.gov/nuccore/NC_000908.2#bakta-gff3"
  origin_record: "NC_000908.2"
  selection: "a Bakta GFF3 with # comments and an embedded ##FASTA section (https://github.com/turbomam/feature-table-corpus/issues/177)"
  retrieved: "2026-11-06"
  license: "CC0 for this output; Bakta database 6.0 is CC-BY-4.0 (https://doi.org/10.5281/zenodo.14916843)"
  notes: "Produced in this repository on 2026-11-06 by scripts/annotate_with_bakta_and_prokka.py from the NCBI RefSeq sequence NC_000908.2 (Mycoplasmoides genitalium G37, formerly Mycoplasma genitalium, the name still used in the file names here; 580,076 bp, translation table 4), run in Docker on an Apple M5 Max under linux/amd64 emulation. Bakta v1.12.2 (oschwengers/bakta:v1.12.2) with database 6.0 light (https://doi.org/10.5281/zenodo.14916843, md5 4a6e059ded39e9c5537ef4137d2f5648) after amrfinder_update; --keep-contig-headers keeps the NCBI sequence name. The light database leaves more proteins as hypothetical than the full one."
- id: bakta-mgenitalium-g37-genbank
  label: "Bakta GenBank, Mycoplasmoides genitalium G37 (NC_000908.2)"
  tier: "vendored"
  path: "corpus/sources/bakta/Mycoplasma_genitalium_G37_NC_000908.2_bakta-1.12.2.gbff"
  format: "GenBank"
  source: "Bakta v1.12.2 run on NCBI RefSeq NC_000908.2"
  producer: "Bakta 1.12.2"
  producer_kind: "annotation pipeline"
  bytes: 1299978
  md5: "8058d787b5a72056b1d5b22872512a78"
  sha256: "5bf8027f00f574c67632b73cd5600fd657e203ceffbe2b78bbd6fa53beb1d3be"
  origin_url: "https://www.ncbi.nlm.nih.gov/nuccore/NC_000908.2#bakta-genbank"
  origin_record: "NC_000908.2"
  selection: "the same Bakta run as its GFF3, for interconversion"
  retrieved: "2026-11-06"
  license: "CC0 for this output; Bakta database 6.0 is CC-BY-4.0 (https://doi.org/10.5281/zenodo.14916843)"
  notes: "Produced in this repository on 2026-11-06 by scripts/annotate_with_bakta_and_prokka.py from the NCBI RefSeq sequence NC_000908.2 (Mycoplasmoides genitalium G37, formerly Mycoplasma genitalium, the name still used in the file names here; 580,076 bp, translation table 4), run in Docker on an Apple M5 Max under linux/amd64 emulation. Bakta v1.12.2 (oschwengers/bakta:v1.12.2) with database 6.0 light (https://doi.org/10.5281/zenodo.14916843, md5 4a6e059ded39e9c5537ef4137d2f5648) after amrfinder_update; --keep-contig-headers keeps the NCBI sequence name. The light database leaves more proteins as hypothetical than the full one."
- id: bakta-mgenitalium-g37-tsv
  label: "Bakta TSV feature table, Mycoplasmoides genitalium G37 (NC_000908.2)"
  tier: "vendored"
  path: "corpus/sources/bakta/Mycoplasma_genitalium_G37_NC_000908.2_bakta-1.12.2.tsv"
  format: "TSV"
  source: "Bakta v1.12.2 run on NCBI RefSeq NC_000908.2"
  producer: "Bakta 1.12.2"
  producer_kind: "annotation pipeline"
  bytes: 51970
  md5: "f9fc5b28c6ace3b3f25bc6c9e8943bbb"
  sha256: "8b17e4bba562135d1faa6357fbda94a9889044c9671117655677b1788c5b63ac"
  origin_url: "https://www.ncbi.nlm.nih.gov/nuccore/NC_000908.2#bakta-tsv"
  origin_record: "NC_000908.2"
  selection: "the same Bakta run as its GFF3, for interconversion"
  retrieved: "2026-11-06"
  license: "CC0 for this output; Bakta database 6.0 is CC-BY-4.0 (https://doi.org/10.5281/zenodo.14916843)"
  notes: "Produced in this repository on 2026-11-06 by scripts/annotate_with_bakta_and_prokka.py from the NCBI RefSeq sequence NC_000908.2 (Mycoplasmoides genitalium G37, formerly Mycoplasma genitalium, the name still used in the file names here; 580,076 bp, translation table 4), run in Docker on an Apple M5 Max under linux/amd64 emulation. Bakta v1.12.2 (oschwengers/bakta:v1.12.2) with database 6.0 light (https://doi.org/10.5281/zenodo.14916843, md5 4a6e059ded39e9c5537ef4137d2f5648) after amrfinder_update; --keep-contig-headers keeps the NCBI sequence name. The light database leaves more proteins as hypothetical than the full one."
- id: prokka-mgenitalium-g37-gff3
  label: "Prokka GFF3, Mycoplasmoides genitalium G37 (NC_000908.2)"
  tier: "vendored"
  path: "corpus/sources/prokka/Mycoplasma_genitalium_G37_NC_000908.2_prokka-1.15.6.gff"
  format: "GFF3"
  source: "Prokka v1.15.6 run on NCBI RefSeq NC_000908.2"
  producer: "Prokka 1.15.6"
  producer_kind: "annotation pipeline"
  bytes: 707621
  md5: "b6a106fa636882550c82ccb422312105"
  sha256: "300ede270f5cae0750bcb05e79a8667dd5d8b40d969d38f36418fdbddb970122"
  origin_url: "https://www.ncbi.nlm.nih.gov/nuccore/NC_000908.2#prokka-gff3"
  origin_record: "NC_000908.2"
  selection: "a Prokka GFF3 with an embedded ##FASTA section, on the same genome as the Bakta files"
  retrieved: "2026-11-06"
  license: "CC0 for this output"
  notes: "Produced in this repository on 2026-11-06 by scripts/annotate_with_bakta_and_prokka.py from the NCBI RefSeq sequence NC_000908.2 (Mycoplasmoides genitalium G37, formerly Mycoplasma genitalium, the name still used in the file names here; 580,076 bp, translation table 4), run in Docker on an Apple M5 Max under linux/amd64 emulation. Prokka v1.15.6 (staphb/prokka:1.15.6) with --gcode 4."
- id: prokka-mgenitalium-g37-genbank
  label: "Prokka GenBank, Mycoplasmoides genitalium G37 (NC_000908.2)"
  tier: "vendored"
  path: "corpus/sources/prokka/Mycoplasma_genitalium_G37_NC_000908.2_prokka-1.15.6.gbk"
  format: "GenBank"
  source: "Prokka v1.15.6 run on NCBI RefSeq NC_000908.2"
  producer: "Prokka 1.15.6"
  producer_kind: "annotation pipeline"
  bytes: 1216079
  md5: "38f3ba4c65fbb635b72bdeec9710daa7"
  sha256: "1b3563c82cc30bf985242abd172adec4533add316fe6927dd0d6418356e67cbb"
  origin_url: "https://www.ncbi.nlm.nih.gov/nuccore/NC_000908.2#prokka-genbank"
  origin_record: "NC_000908.2"
  selection: "the same Prokka run as its GFF3, for interconversion"
  retrieved: "2026-11-06"
  license: "CC0 for this output"
  notes: "Produced in this repository on 2026-11-06 by scripts/annotate_with_bakta_and_prokka.py from the NCBI RefSeq sequence NC_000908.2 (Mycoplasmoides genitalium G37, formerly Mycoplasma genitalium, the name still used in the file names here; 580,076 bp, translation table 4), run in Docker on an Apple M5 Max under linux/amd64 emulation. Prokka v1.15.6 (staphb/prokka:1.15.6) with --gcode 4."
- id: prokka-mgenitalium-g37-tbl
  label: "Prokka feature table (.tbl), Mycoplasmoides genitalium G37 (NC_000908.2)"
  tier: "vendored"
  path: "corpus/sources/prokka/Mycoplasma_genitalium_G37_NC_000908.2_prokka-1.15.6.tbl"
  format: "NCBI feature table"
  source: "Prokka v1.15.6 run on NCBI RefSeq NC_000908.2"
  producer: "Prokka 1.15.6"
  producer_kind: "annotation pipeline"
  bytes: 101644
  md5: "30a831c87454b46f8d2554b35708370c"
  sha256: "a223c20fd938d4ef65f125feff96e3f8720024a97175c9f43a28819def9e107f"
  origin_url: "https://www.ncbi.nlm.nih.gov/nuccore/NC_000908.2#prokka-tbl"
  origin_record: "NC_000908.2"
  selection: "the same Prokka run as its GFF3, for interconversion"
  retrieved: "2026-11-06"
  license: "CC0 for this output"
  notes: "Produced in this repository on 2026-11-06 by scripts/annotate_with_bakta_and_prokka.py from the NCBI RefSeq sequence NC_000908.2 (Mycoplasmoides genitalium G37, formerly Mycoplasma genitalium, the name still used in the file names here; 580,076 bp, translation table 4), run in Docker on an Apple M5 Max under linux/amd64 emulation. Prokka v1.15.6 (staphb/prokka:1.15.6) with --gcode 4."
tier_counts:
  vendored: 74
  linked: 24
  restricted: 5
  not located: 1
  derived: 12
