motus classify#

The motus classify command allows users to take unknown genome sequences (in FASTA format) and map them to known taxonomic units (mOTUs) using the mOTUs marker gene database. This is particularly useful for connecting user genomes to their abundances in public or private taxonomic profiles.

Note: This command requires the mOTUs marker gene database to be downloaded. Use motus downloadMGDB to download the database.

1. Preparing Your Input#

The tool does not take FASTA files directly on the command line. Instead, it requires a manifest file: a simple text file containing the paths to your genome files, with one path per line.

Requirements:

  • Files must be in .fa, .fasta, or gzipped .gz format.

  • Crucial: Every filename must be unique, even if they are in different folders.

Download tutorial genomes (4 genome files and 1 manifest file, ~2MB)

$ curl https://zenodo.org/records/19071244/files/genomes.tar.gz -o genomes.tar.gz
$ tar -xzvf genomes.tar.gz
$ ls genomes/

ACIN21-1_SAMN05421555_MAG_00000004.fa
ARTA20-1_SAMN17006390_MAG_00000056.fa
BAEC15-1_SAMEA2580230_MAG_00000003.fa
ROHW24-1_SAMN18246391_MAG_00000126.fa
genomes.txt

$ cat genomes/genomes.txt

genomes/ACIN21-1_SAMN05421555_MAG_00000004.fa
genomes/ARTA20-1_SAMN17006390_MAG_00000056.fa
genomes/BAEC15-1_SAMEA2580230_MAG_00000003.fa
genomes/ROHW24-1_SAMN18246391_MAG_00000126.fa

2. Running motus classify#

To run the classification, use the -i flag for your list and -o for your desired output name.

$ motus classify -i genomes/genomes.txt -o classify_example.tsv -t 32

INFO: mOTU tool starting - mOTUs4:4.1.0
INFO: Starting mOTUs classify:
INFO: Loading database from db_mOTU ...
INFO: mOTUs database version: 4.1
INFO: Loading database finished. Version 4.1 (version date: 2026-01-26) contains 124300 mOTUs, 2075157 markergeneclusters and 3436253 markergenes.
INFO:         Input = 4 genomes.
INFO:         Output will be written to classify_example.tsv
INFO:         Temporary files will be written to classify_example.tsv_classify_tmp
INFO:         Running fetchMGs on genomes.
INFO: Starting gene calling from 4 genome files.            100%|██████████ 4/4 [00:04<00:00,  1.11s/genomes]
INFO: Finished gene calling.
INFO: Starting marker gene extraction from 4 protein files. 100%|██████████ 4/4 [00:02<00:00,  1.68protein files/s]
INFO: Finished marker gene extraction.
INFO:         Finished running fetchMGs on genomes.
INFO:         Collecting fetchMGs results
INFO:         Finished collecting fetchMGs results. Genomes = 4, Genomes with enough MGs = 3
INFO:         Matching genomes against the mOTUs database
INFO:         Aligning genome marker genes against the mOTUs marker gene database using vsearch
INFO:         Finished alignment
INFO:         Combining individual marker gene distances into genome to genome distances
INFO:         Finished combining
INFO: mOTU tool shutting down with exitcode 0

Understanding the Options:

  • -i: Path to your text file listing the genomes.

  • -o: The name of the resulting classification file.

  • -t: Number of threads.

Note: Some parts of the classify routine could be parallelized. Others couldn’t. In general it’s recommended to scale up to 32 cores.

3. Understanding the Output#

The tabular output file contains one line per submitted genome (see Classify file) with the following columns:

  • GENOME: the submitted genome.

  • CLOSEST_MOTU: the most similar mOTU found, or no_mOTU_<6MGs if fewer than 6 of the 10 marker genes were detected (e.g. ACIN21-1_SAMN05421555_MAG_00000004.fa), or no_mOTU if enough marker genes were present but no mOTU could be matched at all.

  • SIMILARITY: similarity (%) to the closest mOTU, or -1.0 when no match was found.

  • ASSIGNED_TO_MOTU: True if the genome was assigned to CLOSEST_MOTU (similarity ≥ 96.5%), otherwise False.

  • TAXONOMY: the GTDB taxonomy of the closest mOTU (empty ranks d__;p__;c__;o__;f__;g__;s__ when no match was found).

  • #MGs: the number of marker genes detected in the genome.

Note: Since v4.1, the output always reports the closest mOTU together with a separate ASSIGNED_TO_MOTU flag. A genome that has at least six marker genes but whose closest mOTU falls below the 96.5% cutoff (reported as Novel-no_mOTU in earlier versions) now lists that closest mOTU and its actual similarity and indicates False in the ASSIGNED_TO_MOTU column. In the example below, ROHW24-1_SAMN18246391_MAG_00000126.fa is 86.6% similar to mOTUv4.0_025074 and is therefore reported but not assigned.

$ cat classify_example.tsv

GENOME                                 CLOSEST_MOTU     SIMILARITY  ASSIGNED_TO_MOTU  TAXONOMY                                                                                                                                     #MGs
ACIN21-1_SAMN05421555_MAG_00000004.fa  no_mOTU_<6MGs          -1.0  False             d__;p__;c__;o__;f__;g__;s__                                                                                                                     1
ARTA20-1_SAMN17006390_MAG_00000056.fa  mOTUv4.0_000062       99.62  True              d__Bacteria;p__Actinomycetota;c__Coriobacteriia;o__Coriobacteriales;f__Eggerthellaceae;g__Ellagibacter;s__Ellagibacter isourolithinifaciens     8
BAEC15-1_SAMEA2580230_MAG_00000003.fa  mOTUv4.0_007086       99.71  True              d__Bacteria;p__Bacillota;c__Bacilli;o__Erysipelotrichales;f__Erysipelotrichaceae;g__Faecalicoccus;s__Faecalicoccus sp900546545                  6
ROHW24-1_SAMN18246391_MAG_00000126.fa  mOTUv4.0_025074        86.6  False             d__Bacteria;p__Actinomycetota;c__Actinomycetes;o__Nanopelagicales;f__Nanopelagicaceae;g__Planktophila;s__Planktophila sp036734125               8

ico1 mOTUs is part of SIB's portfolio of open tools and databases.

ico2 mOTUs is part of the ELIXIR-CH Service Delivery Plan.