ncbi-viruses-refseq-2025.01.k31-sc50_0.9995_pretrained.zip (for YACHT >= 2.0.0)

For YACHT 2.0.0 or later; YACHT 1.x cannot use this database. YACHT pretrained reference database ncbi-viruses-refseq-2025.01.k31-sc50_0.9995: NCBI RefSeq viral genomes (January 2025), at scaled 50: the curated subset of the NCBI viral assemblies. 16081 genomes (from 16279 genome sketches; ANI threshold 0.9995). Download with yacht download pretrained_ref_db --name ncbi-viruses-refseq-2025.01.k31-sc50 --ani_thresh 0.9995. Sources, licenses, identifiers, how coarse the database is and how to retrieve the genomes are in the data card README.md (in this record and inside the zip).