FASTag

Partial sequence tags from peptide MS/MS spectra, and a filter that keeps only the spectra whose tags occur in sequences you supply.

Download GitHub Documentation

48×

faster than DirecTag on PXD047793 (3.09 M spectra)

280 MB

peak memory on a 23 GB run (DirecTag: 69.4 GB)

98.6%

of identified spectra get a valid tag

5

platforms, macOS signed & notarized

An OpenMS TOPP tool — reads mzML and mzPeak, writes tab-separated tags, and drops into an OpenMS pipeline like any other.

What it does

A sequence tag is a short peptide read inferred directly from a fragment-ion ladder — no database search, no precursor digest. FASTag is a fast, sensitive prefilter: it tells you what is worth looking at before you commit to a slow method.

• Narrow a huge search space to the spectra carrying a tag from your protein(s) of interest, with -fasta.
• Get a sense of what is in a run — species, contaminants, a specific protein — without a full database search.
• Feed tags into a downstream tool: open-modification search, spectral clustering, de novo seeding.

The algorithm is a reimplementation of DirecTag (Tabb et al., J. Proteome Res. 2008, 7:3838). That paper is the citation — FASTag has no separate publication.

Why use it

Measured against DirecTag 1.4.21142 at 128 threads on public ProteomeXchange runs, with a September 2026 FASTag build and settings matched to DirecTag (tag length 3, 50 tags, 100 peaks, no gaps, no deisotoping); wall clock and peak memory for the whole run, including writing the tags:

Memory is flat in the file size rather than proportional to it, so a run is bounded by threads, not by input. Both tools reach the same accuracy ceiling on PXD000001 — 98.62% of the spectra identified by its Mascot search get a valid tag — and FASTag gets there sooner in the ranking. A benchmark of the current release on further public data, at several thread counts and against database-search identifications, is in preparation.

Download

Binaries are self-contained — the tool, its libraries and the OpenMS data it needs. macOS builds are signed with a Developer ID certificate, notarized by Apple, and the disk images carry their notarization ticket, so they open with no network connection.

Species detection additionally needs the taxonomy index, FASTag-taxonomy-k7.tar.gz, which is already embedded in the release tarballs above. Windows binaries are not yet Authenticode-signed, so SmartScreen will warn on first run. All releases: github.com/okohlbacher/FASTag/releases.

Quick start

# tags for every MS2 spectrum in a run
FASTag -in run.mzML -out tags.tsv

# high-resolution data, seed length 3 extended to at most 15 residues
FASTag -in run.mzML -out tags.tsv -tag_length 3 -extension 6

# ion-trap MS2 — set the tolerance to match the analyser
FASTag -in run.mzML -out tags.tsv -fragment_tolerance 0.3 -fragment_tolerance_unit Da

# keep only tags occurring in a protein of interest, and the spectra carrying them
FASTag -in run.mzML -out tags.tsv -fasta protein.fasta -out_spectra hits.mzML

# mzPeak in, mzPeak out — any combination of mzML and mzpeak works
FASTag -in run.mzpeak -out tags.tsv -out_spectra hits.mzpeak

Isotope collapsing and one gap per tag are on by default. For the fastest possible run, or to match a tool with no equivalent of either, add -no_deisotope -gaps 0.

mzPeak

FASTag reads and writes mzPeak, the Parquet-backed mass-spectrometry container. Tagging is unaffected by which you use: reading a run as mzPeak gives the same spectra as reading it as mzML, and a run written to mzPeak and tagged again reproduces the original tag set exactly.

• The same tags from mzML and from both mzPeak layouts (PXD003531, 27,345 MS2 spectra).
• Profile MS2 is centroided on read, which recovers tags a raw profile read loses.

Support comes from mzpeak-openms, bundled with every released binary.

Documentation

• README — options, output format, reconciliation, entrapment-calibrated confidence, species detection, streaming mode
• BENCHMARK-DirecTag.md — accuracy and throughput against the reference implementation on identified spectra
• TEST-DATA.md — what each benchmark file is and where it came from
• Issues — bugs and questions

Citation

For the algorithm:

Tabb DL, Ma ZQ, Martin DB, Ham AJ, Chambers MC. DirecTag: accurate sequence tags from peptide MS/MS through statistical scoring. J Proteome Res 2008; 7(9): 3838–46. doi:10.1021/pr800154p

For the framework:

Pfeuffer J, Bielow C, Wein S, et al. OpenMS 3 enables reproducible analysis of large-scale mass spectrometry data. Nat Methods 2024. doi:10.1038/s41592-024-02197-7