merge.yaml

---
# Standard merge configuration (Bakta and COG excluded)
# For Bakta-enabled merge, use merge_bakta.yaml instead
#
# Usage: poetry run kg merge -y merge.yaml
#
configuration:
  output_directory: data/merged
  checkpoint: false
  # Allow domain-specific categories from METPO ontology
  # METPO:1004005 represents the "growth medium" ontology class
  # This preserves semantic precision for microbial growth media
  category_allowlist:
    - "METPO:1004005"  # growth medium
  curie_map:
    # define non-canonical CURIE to IRI mappings (for RDF)
  node_properties:
    # define predicates that are to be treated as direct node properties (for RDF)
  predicate_mappings:
    # map non-canonical predicates to a property name (for RDF)
  property_types:
    # define the type for non-canonical properties for RDF export
  preserve:
    - primary_knowledge_source

merged_graph:
  name: kg-microbe graph
  source:
    ncbitaxon:
      name: "NCBITaxon"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies/ncbitaxon_nodes.tsv
          - data/transformed/ontologies/ncbitaxon_edges.tsv
    chebi:
      name: "CHEBI"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies/chebi_nodes.tsv
          - data/transformed/ontologies/chebi_edges.tsv
    envo:
      name: "ENVO"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies/envo_nodes.tsv
          - data/transformed/ontologies/envo_edges.tsv
    go:
      name: "GO"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies/go_nodes.tsv
          - data/transformed/ontologies/go_edges.tsv
    # mondo:
    #   name: "MONDO"
    #   input:
    #     format: tsv
    #     filename:
    #       - data/transformed/ontologies/mondo_nodes.tsv
    #       - data/transformed/ontologies/mondo_edges.tsv
    # hp:
    #   name: "HP"
    #   input:
    #     format: tsv
    #     filename:
    #       - data/transformed/ontologies/hp_nodes.tsv
    #       - data/transformed/ontologies/hp_edges.tsv
    ec:
      name: "EC"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies/ec_nodes.tsv
          - data/transformed/ontologies/ec_edges.tsv
    metpo:
      name: "METPO"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies/metpo_nodes.tsv
          - data/transformed/ontologies/metpo_edges.tsv
    # Selective per-CURIE stub-import for NCIT, mesh, BTO, PO, and MICRO —
    # only the IDs referenced by mappings/* are imported (not the full
    # ontologies). See kg_microbe/transform_utils/ontologies_stubs/ for
    # the transform.
    ontologies_stubs:
      name: "ontologies_stubs"
      input:
        format: tsv
        filename:
          - data/transformed/ontologies_stubs/ncit_nodes.tsv
          - data/transformed/ontologies_stubs/ncit_edges.tsv
          - data/transformed/ontologies_stubs/mesh_nodes.tsv
          - data/transformed/ontologies_stubs/mesh_edges.tsv
          - data/transformed/ontologies_stubs/bto_nodes.tsv
          - data/transformed/ontologies_stubs/po_nodes.tsv
          - data/transformed/ontologies_stubs/po_edges.tsv
          - data/transformed/ontologies_stubs/micro_nodes.tsv
          - data/transformed/ontologies_stubs/micro_edges.tsv
    bacdive:
      name: "bacdive"
      input:
        format: tsv
        filename:
          - data/transformed/bacdive/nodes.tsv
          - data/transformed/bacdive/edges.tsv
    mediadive:
      name: "mediadive"
      input:
        format: tsv
        filename:
          - data/transformed/mediadive/nodes.tsv
          - data/transformed/mediadive/edges.tsv
    rhea_mappings:
      name: "rhea_mappings"
      input:
        format: tsv
        filename:
          - data/transformed/rhea_mappings/nodes.tsv
          - data/transformed/rhea_mappings/edges.tsv
    upa:
      input:
        name: "upa"
        format: tsv
        filename:
          - data/transformed/ontologies/upa_nodes.tsv
          - data/transformed/ontologies/upa_edges.tsv
    madin_etal:
      input:
        name: "madin_etal"
        format: tsv
        filename:
          - data/transformed/madin_etal/nodes.tsv
          - data/transformed/madin_etal/edges.tsv
    metatraits:
      input:
        name: "metatraits"
        format: tsv
        filename:
          - data/transformed/metatraits/nodes.tsv
          - data/transformed/metatraits/edges.tsv
    metatraits_gtdb:
      input:
        name: "metatraits_gtdb"
        format: tsv
        filename:
          - data/transformed/metatraits_gtdb/nodes.tsv
          - data/transformed/metatraits_gtdb/edges.tsv
    bactotraits:
      input:
        name: "bactotraits"
        format: tsv
        filename:
          - data/transformed/bactotraits/nodes.tsv
          - data/transformed/bactotraits/edges.tsv
    # bakta_cmm:
    #   name: "bakta_cmm"
    #   input:
    #     format: tsv
    #     filename:
    #       - data/transformed/bakta/cmm_bakta/nodes.tsv
    #       - data/transformed/bakta/cmm_bakta/edges.tsv
    # bakta_pfas:
    #   name: "bakta_pfas"
    #   input:
    #     format: tsv
    #     filename:
    #       - data/transformed/bakta/pfas_bakta/nodes.tsv
    #       - data/transformed/bakta/pfas_bakta/edges.tsv
    # cog:
    #   name: "cog"
    #   input:
    #     format: tsv
    #     filename:
    #       - data/transformed/cog/nodes.tsv
    #       - data/transformed/cog/edges.tsv
    gtdb:
      name: "GTDB Taxonomy"
      input:
        format: tsv
        filename:
          - data/transformed/gtdb/nodes.tsv
          - data/transformed/gtdb/edges.tsv
    # LPSN nomenclature. `lpsn` is the fast GSS/CSV base layer (taxon
    # names, ranks, GTDB/NCBITaxon cross-refs); `lpsn_api` is the
    # authenticated JSON-API enrichment (above-genus taxonomy, basonyms,
    # publication DOIs/PMIDs, 16S INSDC accessions). The API enrichment
    # nodes reference lpsn:* records minted by the GSS layer, so both are
    # merged together.
    lpsn:
      name: "LPSN"
      input:
        format: tsv
        filename:
          - data/transformed/lpsn/nodes.tsv
          - data/transformed/lpsn/edges.tsv
    lpsn_api:
      name: "LPSN API"
      input:
        format: tsv
        filename:
          - data/transformed/lpsn_api/nodes.tsv
          - data/transformed/lpsn_api/edges.tsv
    # kegg:
    #   name: "kegg"
    #   input:
    #     format: tsv
    #     filename:
    #       - data/transformed/kegg/nodes.tsv
    #       - data/transformed/kegg/edges.tsv
    # ctd:
    #   input:
    #     name: "ctd"
    #     format: tsv
    #     filename:
    #       - data/transformed/ctd/nodes.tsv
    #       - data/transformed/ctd/edges.tsv
    # disbiome:
    #   input:
    #     name: "disbiome"
    #     format: tsv
    #     filename:
    #       - data/transformed/disbiome/nodes.tsv
    #       - data/transformed/disbiome/edges.tsv
    # wallen_etal:
    #   input:
    #     name: "wallen_etal"
    #     format: tsv
    #     filename:
    #       - data/transformed/wallen_etal/nodes.tsv
    #       - data/transformed/wallen_etal/edges.tsv
    # Not feasible using kgx merge process
    # uniprot_functional_microbes:
    #   input:
    #     name: "uniprot_functional_microbes"
    #     format: tsv
    #     filename:
    #       - data/transformed/uniprot_functional_microbes/nodes.tsv
    #       - data/transformed/uniprot_functional_microbes/edges.tsv
    # uniprot_human:
    #   input:
    #     name: "uniprot_human"
    #     format: tsv
    #     filename:
    #       - data/transformed/uniprot_human/nodes.tsv
    #       - data/transformed/uniprot_human/edges.tsv
  operations:
    - name: kgx.graph_operations.summarize_graph.generate_graph_stats
      args:
        graph_name: kg-microbe graph
        filename: merged_graph_stats.yaml
        node_facet_properties:
          - provided_by
        edge_facet_properties:
          - provided_by
          - source
  destination:
    merged-kg-tsv:
      format: tsv
      compression: tar.gz
      filename: merged-kg
#    merged-kg-nt:
#      format: nt
#      compression: gz
#      filename: kg_microbe.nt.gz