Skip to main content

UniDecImporter

UniDecImporter is a standalone Python package for reading mass-spectrometry data into NumPy arrays. It was extracted from UniDec, but has no runtime dependency on UniDec or IsoDec.

The same high-level API covers single spectra, LC-MS, CD-MS, and ion-mobility MS:

from UniDecImporter import get_importer

with get_importer("run.mzML") as reader:
    spectrum = reader.get_avg_scan(time_range=(2.0, 2.5))
    tic = reader.get_tic()

Installation

python -m pip install UniDecImporter

For Thermo RAW support on Windows x86-64, install the optional bridge too:

python -m pip install "UniDecImporter[thermo]"

Python 3.10–3.13 is supported on Windows, macOS, and Linux. Open formats work on all three operating systems. The package includes Windows-only Thermo .NET assemblies as data files, but importing and using open-format readers does not load them.

Thermo proprietary software: Installing or using the bundled Thermo RawFileReader assemblies means you accept Thermo's separate license, included in the distribution. End users may not redistribute those assemblies. Commercial exploitation requires Thermo's prior written consent. The BSD license covers this project's Python code, not the Thermo binaries. See THERMO_RAWFILEREADER_TERMS.md.

RawFileReader reading tool. Copyright © 2016 by Thermo Fisher Scientific, Inc. All rights reserved.

Format support

Format Data types Platforms Reader dependency
mzML, indexed mzML.gz LC-MS, MS/MS, CD-MS, IM-MS Windows, macOS, Linux pymzML
mzXML LC-MS, MS/MS, CD-MS Windows, macOS, Linux pyteomics
TXT, DAT, CSV Single scan, CD-MS, IM-MS Windows, macOS, Linux NumPy
NPZ, BIN Single scan, CD-MS Windows, macOS, Linux NumPy
I2MS, DMT CD-MS Windows, macOS, Linux Python sqlite3
Thermo RAW LC-MS, MS/MS, CD-MS Windows x86-64 only Thermo RawFileReader + pythonnet
Waters RAW directory LC-MS, IM-MS Windows x86-64 only Waters MassLynx SDK
Agilent .d LC-MS, MS/MS Windows x86-64 only Agilent MassHunter Data Access SDK + pythonnet

NPZ single-scan files must contain an array named data. Numeric arrays load without pickle support. Legacy CD-MS files with object-backed data arrays are also supported, but emit a warning because they require pickle deserialization; load those files only from trusted sources.

The four Thermo RawFileReader assemblies are bundled under their separate proprietary terms. Waters and Agilent SDK licenses do not permit this project to redistribute those binaries, so users must obtain them separately. Set THERMO_RAW_FILE_READER_DIR to override the bundled Thermo assembly directory, MASSLYNX_RAW_DLL to a licensed Waters DLL, or AGILENT_DA_SDK_DIR to a licensed Agilent Data Access assembly directory. Unsupported platforms and missing runtimes raise VendorReaderUnavailableError with an actionable message.

Core API

get_importer(file_path, **options) inspects the path and returns the appropriate reader. file_path may be a string or path-like object; options are passed to the format-specific reader. Unknown extensions raise UnsupportedFormatError, while an unavailable proprietary SDK or platform raises VendorReaderUnavailableError.

All readers inherit from UniDecImporter.Importer.Importer. They expose scans (ordered scan identifiers), times (retention times in minutes), levels (MS order when available), and an inclusive scan_range. Ordinary spectra and chromatograms are NumPy N x 2 arrays. Specialized return shapes are noted below.

Spectrum access

Method Description
get_single_scan(scan) Return one spectrum as N x 2 (m/z, intensity).
get_all_scans() Load and return all spectra as a list in scan order.
get_avg_scan(scan_range=None, time_range=None, sum_mode=False) Average spectra over an inclusive scan or retention-time range; set sum_mode=True to sum intensities.
avg_fast(scan_range=None, time_range=None, sum_mode=False) Average spectra through the shared cached-data implementation. Most applications should call get_avg_scan.
get_mz_localmax(mz, mz_tol) Find each scan's local peak near mz; mz_tol is in ppm. Return (m/z, intensity) rows and omit scans with no peak.

Supplying time_range selects the nearest endpoint scans and takes precedence over a simultaneously supplied scan_range. With neither argument, averaging uses the full reader range.

Scan metadata and coordinate conversion

Method Description
get_polarity(scan=None) Return "Positive", "Negative", or a reader-specific unknown value.
get_ms_order(scan=1) Return the scan's MS level; readers without level metadata default to 1.
get_max_scan() Return the final scan identifier.
get_max_time() Return the final retention time in minutes.
get_scan_index(scan) Convert a scan identifier to its array index. Identifiers outside the available range clamp to the first or last index.
get_scan_time(scan) Return the retention time for a scan identifier.
get_time_scan(time) Return the scan identifier nearest to a retention time in minutes.
get_scans_from_times(time_range) Convert two retention times to an inclusive pair of scan identifiers.
get_times_from_scans(scan_range) Return [start, midpoint, end] retention times for an inclusive scan range.
scan_range_from_inputs(scan_range=None, time_range=None) Resolve either selection form and clamp it to the available scans. This is primarily useful to reader implementers.
check_centroided() Estimate centroid/profile status from the first selected scan using a lag-one autocorrelation heuristic and update reader.centroided.

Chromatograms

These methods require reader.chrom_support:

Method Description
get_tic() Return the total-ion chromatogram as N x 2 (retention time, intensity).
index_scans(min_mz, bin_width) Build or replace the in-memory MS1 peak index used for extracted-ion chromatograms.
get_eic(mass, mz_tol, rt_range=None) Return an extracted-ion chromatogram around mass; mz_tol is an absolute m/z tolerance and rt_range is optional. The default index is built automatically on first use.

CD-MS and ion mobility

Check reader.cdms_support and reader.imms_support before calling these methods:

Method Description
get_cdms_data(scan_range=None) Return an N x 5 array: m/z, intensity, scan, inverse injection time, and retention time. Some readers override the base metadata defaults.
get_imms_scan(scan) Return one ion-mobility scan as N x 3 (m/z, drift time, intensity).
get_all_imms_scans() Load all ion-mobility scans in scan order.
get_imms_avg_scan(scan_range=None, time_range=None, mzbins=1) Merge selected mobility scans into an N x 3 array. mzbins sets the linear m/z bin width; a false-like value requests automatic spacing.

Calling an unsupported specialized method raises NotImplementedError.

Construction and resource management

Importer.__init__(file_path, **kwargs) initializes the shared state, but applications should construct readers through get_importer. close() releases open files, database connections, or vendor handles; the base implementation is a no-op. __enter__() and __exit__() provide context-manager support, call close() on exit, and do not suppress exceptions. Prefer with whenever practical:

with get_importer("run.mzML") as reader:
    data = reader.get_avg_scan()

For a two-column single spectrum:

from UniDecImporter import get_importer

reader = get_importer("spectrum.csv")
data = reader.get_avg_scan()
assert data.shape[1] == 2

For CD-MS:

with get_importer("ions.dmt") as reader:
    events = reader.get_cdms_data()
    mz, intensity, scan, inverse_injection_time, time = events.T

Development

git clone https://github.com/michaelmarty/UniDecImporter.git
cd UniDecImporter
python -m pip install -e ".[test]"
python -m pytest

The suite includes fast numerical/unit tests, cross-platform integration tests against the bundled open formats, and separately marked Windows vendor tests:

python -m pytest -m "not integration and not vendor"
python -m pytest -m "integration and not vendor"
python -m pytest -m vendor

See the documentation, PUBLISHING.md, and THIRD_PARTY_NOTICES.md for full details.

Change Log

Version 1.0.2

Well, we got too fancy and burned up a bunch of GitHub LFS bandwidth, so this version uses a smaller set of test files and removes Git LFS from the repository.

Improved test coverage for mzML, mzXML, and NPZ readers.

Harmonized sum vs average behavior to have average by default and sum only when sum_mode=True.

Version 1.0.1

Fix to npz bug.

Version 1.0.0

Initial release of standalone package.

Metadata

Release files for UniDecImporter 1.0.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for UniDecImporter 1.0.2
File Size Uploaded
unidecimporter-1.0.2.tar.gz 489.8 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for UniDecImporter 1.0.2
File Interpreter ABI Platform
unidecimporter-1.0.2-py3-none-any.whl Python 3 none any Details

Total release size: 973.8 kB

Release files / unidecimporter-1.0.2.tar.gz

Download URL unidecimporter-1.0.2.tar.gz
Size 489.8 kB
Tags Source
SHA-256 checksum
How to use checksums
9226df35b91f0570ce99916d6cf8a7ba8ba8c0bb7524c15b3139fb01c648846c
BLAKE2b-256 checksum
How to use checksums
e590ad4b18b1e9c4c08988cbb6d20348be5c53e66754c978bb3b3a85ca611083
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 21, 2026.

Transparency log

Release files / unidecimporter-1.0.2-py3-none-any.whl

Download URL unidecimporter-1.0.2-py3-none-any.whl
Size 484.0 kB
Tags Python 3
SHA-256 checksum
How to use checksums
23a9145e61799255162fe7773ab011f413207d293b0fdec0e23903c89a48de8b
BLAKE2b-256 checksum
How to use checksums
aacdaf0d12056428b22be0dd4dbf2424311945c9e6e7ba61e90478182c33fffd
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Sep 21, 2026.

Transparency log

Release history Release notifications | RSS feed

1.0.3

2 release files

This release

1.0.2 This release

2 release files

1.0.1

2 release files

1.0.0

2 release files

0.1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page