Skip to content

Latest commit

 

History

History
28 lines (23 loc) · 1.79 KB

QC.md

File metadata and controls

28 lines (23 loc) · 1.79 KB
Raw

QC

all filenames must contain KEY4:

gensp.strain.gnm.[.ann].KEY4.[...]

YAML in general

  • do not use tab characters anywhere
  • place quotes around values containing ":" and other YAML-relevant characters, including line breaks
  • separate documents within a file (e.g. strains descriptions file) with ---, not ### or something else

README.yaml:

  • fields may be none other than: scientific_name_abbrev, license, publication_title, taxid, geoseries, expression_unit, dataset_release_date, related_to, bioproject, data_curators, synopsis, genotype, source, local_file_creation_date, provenance, identifier, keywords, publication_doi, citations, genbank_accession, dataset_doi, contributors, original_file_creation_date, description, public_access_level, scientific_name, genotyping_method, sraproject, genotyping_platform
  • identifier must be full name of collection (e.g. PI483463.gnm1.ann1.3Q3Q, not just 3Q3Q)
  • content shared between READMEs must be identical: for example, if a publication is in a genomic README and an annotation README, the publication titles must match exactly.

gene_models_main.gff3

  • must have ID and/or Name attributes in every record
    1. If ONLY ID present, then it must be an LIS conformant ID: gensp.strain.gnm.ann.ShortName.
    2. If Name is also present, then it can be ShortName and the mine loader will ignore ID and prepend the full yuck to Name for the primaryIdentifier.
    3. A third option is to set primaryIdentifier=ID and secondaryIdentifier=Name with a loader option.
  • IDs must match IDs in corresponding FASTA files (.transcript, .cds)

transcript FASTA files

  • IDs match IDs in GFFs
  • use transcript, not mrna, filename suffix
    • WRONG: glyso.F_IGA1003.gnm1.ann1.G61B.mrna.fna.gz
    • CORRECT: glyso.F_IGA1003.gnm1.ann1.G61B.transcript.fna.gz