Detection of species at risk in environmental DNA / shotgun metagenomics is bottlenecked by the same thing that bottlenecks AIS detection: missing or fragmentary reference genomes for the species that matter. SARGDB is the SAR-side companion to AISGDB — same tier scheme, same NCBI pipeline, focused on the species lists set out in federal and Manitoba conservation legislation.
Companion pipeline: danaSeq's nanopore_live, whose --mapping_refs module consumes any reference directory conforming to the SARGDB/AISGDB schema; the SPA exposes a dedicated /sar view alongside /ais.
All 670 species on the federal Species at Risk Act Schedule 1, tiered against the NCBI Datasets v2 + Entrez APIs (assembled WGS / unassembled WGS via SRA / transcriptome / markers / nothing). Filterable + sortable by status (Extirpated / Endangered / Threatened / Special Concern) and taxon group. Outputs: TSV · JSON.
Combined provincial + MB-relevant federal listings — 79 species total: the 65 on MB Regulation 25/98 (Threatened, Endangered and Extirpated Species Regulation) plus 14 SARA Schedule 1 species whose population qualifier names an MB-relevant geography (Manitoba, Saskatchewan, Hudson Bay, Prairie, Boreal, …). MB ESEA itself contains no fish; pulling in SARA-population brings in lake sturgeon, bigmouth buffalo, bull trout, etc. with their Manitoba populations. Each row carries an mb_source column (ESEA / SARA / both). Outputs: TSV · JSON.
The inventory scripts are fully reproducible and re-runnable as new assemblies land:
# re-parse SARA Schedule 1 from the vendored HTML (only needed on a new consolidation)
python3 data/sara_schedule1.parse.py
# query NCBI for the federal SAR list (cache + retry-safe)
python3 canada_sara_inventory.py
# query NCBI for the MB ESEA list (shares the cache; joins overlap)
python3 manitoba_sar_inventory.py