openszraw 0.1.0

Python bindings for the OpenSZRaw Shimadzu LabSolutions .lcd / .qgd file parser

OpenSZRaw

Rust and Python reader for Shimadzu LabSolutions mass spectrometry raw data (.lcd LC-MS, .qgd GCMS, .gcd GC), clean-room reverse-engineered with no Shimadzu SDK or software dependency.

Sibling readers in the same stack: OpenTFRaw (Thermo), OpenWRaw (Waters), OpenTimsTDF (Bruker), OpenARaw (Agilent), OpenSXRaw (SCIEX).

Status

A Rust reader (crates/openszraw) implements all three confirmed raw data variants: .qgd GC-MS (full-scan profile and MRM/targeted), .lcd IT-TOF (run-length-encoded profile spectra, calibrated to physical m/z via the file's own embedded TOF tuning data), and .lcd QTOF (centroid). See docs/format/ for the byte-level format specs and docs/format/06-known-limitations.md for what is deliberately not yet resolved (per-channel polarity, some MS2 precursor m/z values). Python bindings (crates/openszraw-py) mirror the Rust API; neither the Rust crate nor the Python package has been published (crates.io / PyPI) yet. Not yet wired into openmassspec-io as a shimadzu feature. See the sourcing strategy in the ops repo's SCOPING_PLAN.md and this repo's re/ROADMAP.md (local-only, gitignored) for the current phase.

Install (not yet published)

Rust:

cargo add openszraw

Python:

pip install openszraw

Quickstart

Rust:

use openszraw::reader::Reader;
use openmassspec_core::SpectrumSource;

let mut reader = Reader::open("sample.lcd")?;
for spectrum in reader.iter_spectra() {
    println!("{}: {} peaks", spectrum.native_id, spectrum.mz.len());
}

Python:

import openszraw

reader = openszraw.RawReader("sample.lcd")
spectrum = reader.read_spectrum(0)
print(spectrum.ms_level, spectrum.retention_time_sec, len(spectrum.mz))

Reader::open (and RawReader) auto-detects .qgd vs .lcd IT-TOF vs .lcd QTOF from the file's internal CFBF stream layout, never from the filename or extension alone.

License

Apache-2.0. See LICENSE.

The format specification was developed by binary analysis of public mass-spectrometry datasets (PRIDE, MassIVE, and MetaboLights accessions). See CORPUS.md and ATTRIBUTION.md.