Comparison

Comparison with other tools

mashID answers one question quickly: which organism, among those in a database, is this isolate or assembly? Other tools answer neighbouring questions. Use the one that matches yours.

Tool	Method	Best at	Input	Resources	Where mashID differs
mashID	Mash screen (MinHash containment) against a sketch database	Fast species-level identification of isolates, with mixture and coverage flags	Reads or assemblies	Seconds per sample, < 2 GB	Curated names/TaxIDs per database, notes, run provenance, pipeline outputs.
Mash screen	The same containment estimate	The primitive that mashID wraps	Reads or assemblies	Same	mashID adds sample handling, database management, naming, notes and reports.
sourmash	FracMinHash sketches; gather decomposes a sample into references	Metagenome composition, large public databases (GTDB, GenBank)	Reads or assemblies	Minutes; large databases	sourmash gather reports every genome in a mixture with abundances; mashID reports the best hit and flags mixtures.
Kraken2 / Bracken	Exact k-mer to LCA classification of each read	Read-level taxonomic profiles of metagenomes	Reads	Database in RAM (tens of GB for standard)	Per-read classification and full profiles; heavier and slower to set up. mashID has no per-read output.
refseq_masher	Mash against a RefSeq sketch	Same idea as mashID against RefSeq	Reads or assemblies	Light	Fixed database; mashID lets you build and check your own and adds notes and provenance.
KmerFinder	k-mer overlap against a curated database	Web-based identification	Reads or assemblies	Web / CGE tools	Web service; mashID runs locally and offline.
GTDB-Tk	Marker-gene placement in the GTDB reference tree, plus ANI	Authoritative taxonomy of assemblies, novel species	Assemblies only	Hours, > 60 GB RAM, 100 GB database	The most rigorous classification; not for reads or quick checks.
FastANI / skani	Average nucleotide identity between genomes	Confirming species boundaries between two assemblies	Assemblies	Light	Pairwise ANI, not a database search; a good second step after mashID.
MLST / cgMLST, SNP typing	Allele or SNP calling against a scheme or reference	Within-species typing, lineage, outbreak analysis	Reads or assemblies	Varies	Resolution below the species level, where mashID stops.

When to use what

• Isolate WGS, "what did we sequence?" mashID, with a database for the expected group or a broad one. Confirm surprising results with FastANI against the top reference's assembly.
• Metagenome or complex mixture: sourmash gather or Kraken2 with Bracken; mashID only tells you that a mixture exists and names the strongest components.
• Publication-grade taxonomy of a new assembly: GTDB-Tk.
• Lineage, variant, outbreak: MLST, cgMLST or SNP-based tools after mashID has confirmed the species.