OxoniumBrowserX v2.0.1 (Beta) — Standalone executable version (expanded 27/05/2026)

0. Overview

OxoniumBrowserX enables untargeted exploration of oxonium ions in high resolution (Orbitrap) shotgun proteomics data, with a particular focus on identifying sugar oxonium ions from prokaryotic protein glycosylation involving rare sugars. This standalone executable version includes a graphical user interface (GUI) and does not require a separate Python installation or manual package setup. The current beta version was developed using shotgun proteomics data from various Thermo Orbitrap mass spectrometers (QE, Tribrid, Astral). Raw data from other vendors have not yet been tested. The OxoniumBrowserX workflow:

1. **SAGE database search** — identifies and removes MS2 spectra of unmodified peptides
2. **MS2 spectra recalibration** — two-pass mass calibration using amino acid fragment reference peaks
3. **Oxonium ion detection** — scans remaining spectra for diagnostic sugar ion pairs
4. **Interactive dashboard** — visualize, filter, and export results
5. **Excel reports** — summary and detailed per-scan output files

Note: This is a beta version. Source code, documentation and releases are available at: https://sourceforge.net/projects/oxoniumbrowserx/

1. System requirements

OxoniumBrowserX was tested on 64-bit versions of Windows 10 and Windows 11. The standalone executable package (oxonium_browser_executable_v2.0.1.zip) is available via SourceForge. The software runs locally through the integrated Dash framework and is accessed via a web browser using a local Flask server started automatically by the executable. The software was validated on standard desktop and laptop systems equipped with Intel Core i7 and Intel Xeon processors with 16–64 GB RAM. No dedicated GPU or specialized hardware is required.

Recommended minimum requirements: 4 GB RAM (higher recommended for Astral datasets) 10 GB free disk space (approximately 50 GB recommended for multiple analyses and intermediate files) Both HDD- and SSD-based storage systems were tested successfully.

The interactive dashboard was tested with Mozilla Firefox, Microsoft Edge, and Google Chrome.

2. Installation guide

Download executable package and extract

# Navigate to https://sourceforge.net/projects/oxoniumbrowserx/files/
# Download oxonium_browser_executable_v2.0.1.zip
# Unzip the archive into a local folder

Folder contents

The extracted folder contains:

File	Description
OxBroXv2.0.1.exe	Main standalone executable
Input/	Folder for mzML, fasta and oxonium database files
Output/	Folder where result folders are created
OX_DB_COMBINED_v03.xlsx	Curated and ChemSpace oxonium ion database
README_executable_expanded.md	Documentation

No additional Python installation or package installation is required.

Prepare input files

# Navigate to https://sourceforge.net/projects/oxoniumbrowserx/files/
# Download "TEST_SAMPLE.zip" and unzip
# The folder contains the example mzML, fasta and .xlsx database

Typical installation time

Installation typically requires less than 5 minutes on a standard desktop computer.

3. and 4. Demo, Instructions for use and run on data

After extracting the executable package and preparing the required input files, the software can be started by launching:

OxBroXv2.0.1.exe

The dashboard interface opens automatically in the default web browser.

The demo dataset, MP_14052020_Yeast_AE16_120min_gps_DDA01.mzML, serves as a positive control and primarily contains Hexose and HexNAc oxonium ions derived from high-mannose and oligomannose N- and O-glycans. The dataset originates from a yeast shotgun proteomics experiment performed on a whole-cell lysate without glycopeptide enrichment. Although the collision energies and data acquisition were not optimized for peptide backbone or glycan fragmentation, abundant Hexose and HexNAc oxonium ions are readily observed. The file yeast.fasta contains the Saccharomyces cerevisiae UniProt reference proteome, and OX_DB_COMBINED_v03.xlsx contains curated and theoretical sugar compositions, including ChemSpace-derived entries. Please also check the OxoniumBrowserX wiki at https://sourceforge.net/p/oxoniumbrowserx/wiki/Home/.

Input file summary

Select the following files in the graphical interface:

File	Format	Description
Mass spectrometry data	.mzML	Converted from vendor RAW format (see below)
Protein database	.fasta	Protein sequences for the organism of interest
Sugar oxonium ion database	.xlsx	Provided: OX_DB_COMBINED_v03.xlsx
Output folder	Folder	Directory where results will be stored

The sugar database is included in the download. You only need to provide your .mzML and .fasta files.

Running the analysis

1. Launch the executable
2. The dashboard opens automatically in the browser
3. Select:
4. .mzML file
5. .fasta file
6. oxonium ion .xlsx database
7. output folder
8. Adjust optional parameters if needed:
9. mass error tolerance
10. minimum oxonium intensity
11. amino acid marker requirement
12. Press Run Analysis
13. Wait for processing to complete
14. Explore the interactive dashboard and export results if needed

Converting RAW Files to mzML

Oxonium Browser requires mzML format input. Use ProteoWizard MSConvert (recommended):

1. Open MSConvert GUI and select your RAW file(s)
2. Set output format to **mzML**
3. Recommended settings:
   - Peak Picking: checked (MS levels 1–2; for Astral data, MS level 2 only to reduce size)
   - Binary encoding precision: 64-bit (default); 32-bit acceptable for Astral data to reduce file size
   - zlib compression: checked
   - Write index: checked
   - TPP compatibility: checked
4. Place the resulting `.mzML` file in the desired input location
**Note:** For Astral data, ensure sufficient available memory — the required memory is approximately equal to the mzML file size.

For detailed conversion instructions, see the MSConvert documentation.

Sugar Oxonium Ion Database

Curated database The tool ships with OX_DB_COMBINED_v03.xlsx, a curated database of common, rare, and derivative monosaccharide oxonium ions compiled from literature and databases (CSDB, KEGG), and a set of >3,300 plausible chemical compositions within a defined elemental space (C, H, O, N, S). The empirical entries are marked in the column "list" as "curated" and the chemical space entries as "extensive". Each entry furthermore contains:

Column	Description
Oxonium	Sugar name (e.g. "HexNAc", "Hex", "Hept")
ox_mass1	Primary diagnostic mass (oxonium ion, [M+H−H₂O]⁺)
ox_mass2	Secondary diagnostic mass (water loss, [M+H−2H₂O]⁺)

The database also includes Ox_test_ entries — random mass negative controls for estimating the false match rate. These should not be removed.

Adding custom sugars

To extend the database, add new rows with a unique name in the Oxonium column and the two diagnostic masses. The tool searches for all entries in the file.

Chemspace database (optional use)

For untargeted discovery, OxoniumBrowserX allows searching through >3,300 monosaccharide compositions covering a large space of chemically plausible combinations within a defined elemental space (C, H, O, N, S). These are provided in the OX_DB_COMBINED_v03.xlsx and are marked in the column "list" as "extensive". The dashboard allows switching between curated, and combined views. Thresholds for the combined view are automatically adjusted for the larger search space.

Configurations

The executable version exposes the main user parameters directly in the dashboard interface.

Parameter Reference

MASS_ERROR (Default: 0.001 Da)

Maximum mass error tolerance for matching oxonium ions in MS2 spectra. After processing, check the mass error distribution plot in the dashboard and adjust to match your instrument's actual accuracy. For very high accuracy instruments this can be as low as 0.00025 Da; for older instruments, increase to 0.0025–0.005 Da.

INTENSITY_THRESHOLD (Default: 0.1-0.25%)

Minimum normalized intensity (relative to total spectrum intensity) for oxonium ion detection. The default provides good separation between genuine signals and noise. Can be lowered to 0.1–0.2% for increased sensitivity, or raised to 0.3–0.5% for complex samples with dense background. Monitor the number of random test mass matches when adjusting.

AMINO_ACID_MARKER (Default: False)

When set to True, requires the presence of amino acid marker ions (147.113, 175.119 m/z) alongside sugar oxonium ions. This increases confidence that hits originate from glycopeptides rather than free sugars or other glycoconjugates, at the cost of reduced sensitivity.

SAGE Parameters

The executable version internally uses the following defaults:

Parameter	Default	Description
SAGE_MIN_PEPTIDE_LENGTH	6	Minimum peptide length for database search
SAGE_MISSED_CLEAVAGES	2	Maximum allowed missed cleavages (trypsin KR/P)
SAGE_GENERATE_DECOYS	True	Generate decoy sequences for FDR control
SAGE_PREC_TOL	20	Precursor mass tolerance in ppm
SAGE_FRAG_TOL	20	Fragment mass tolerance in ppm
SAGE_FDR	1	FDR threshold in percent

Expected runtime

For the provided demo dataset, the expected runtime on a standard desktop computer is approximately 2–5 minutes.

Output Files

After a successful run, the following files are generated in the selected output directory:

File	Description
detected_oxoniums_summary.xlsx	Summary of all detected oxonium ions with spectral presence and intensity metrics
detected_oxoniums_detailed.xlsx	Per-scan detection data including retention times
mass_error_two_pass_calibration.png	Mass error distributions before and after calibration
*_psm_validated_sage.tsv	SAGE database search results
*_peptideshaker.sage.tsv	PeptideShaker-compatible SAGE results

Interactive Dashboard

Additionally, the browser-based dashboard provides an interactive view of the results. The user can:

• Filtering controls — adjust count, spectral intensity, and spectral presence thresholds in real time
• Database toggle — switch between curated, or combined views (when chemspace search is enabled)
• Match table — oxonium hits sorted by mass and grouped into ±18 Da water loss families, with color-coded metric columns and export functionality
• Clustered co-occurrence heatmap — pairwise co-occurrence of selected oxonium ions, with rows/columns ordered by Jaccard-based hierarchical clustering and a dendrogram showing grouping structure
• Retention time profiles — extracted ion chromatograms for selected oxonium ions via the checklist
• Mass error distribution — per-reference-peak mass error histograms from the two-pass recalibration
• Export buttons — export the current match table or selected oxonium ion traces to Excel

Troubleshooting

Common Issues Browser window does not open automatically Open the following address manually in a browser:

http://127.0.0.1:8050

Executable immediately closes Ensure all required files are present in the extracted folder and that Windows Defender or antivirus software is not blocking the executable. Dashboard not accessible Verify no other application is already using port 8050. Large mzML files run slowly or crash Ensure sufficient RAM is available. Astral datasets may require memory approximately equal to the mzML file size.

5. License, data privacy, no warranty and contacts

License

OxoniumBrowserX is released under the Apache License 2.0. Copyright (c) 2026 Licensed under the Apache License, Version 2.0 (the "License"); you may not use this software except in compliance with the License. You may obtain a copy of the License at: http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. Please note that some of the functions and libraries used by OxoniumBrowserX may not share the same license as OxoniumBrowserX. If you want to use any of these in a different context, ensure that you obtain the appropriate licenses for the dependent libraries and tools. Dependencies and their licenses: | Library | License | URL | |---------|---------|-----| | Dash | MIT License | https://dash.plotly.com/ | | Pyteomics | MIT License | https://pyteomics.readthedocs.io/ | | pandas | BSD 3-Clause | https://pandas.pydata.org/ | | matplotlib | PSF License | https://matplotlib.org/ | | scipy | BSD 3-Clause | https://scipy.org/ | | numpy | BSD 3-Clause | https://numpy.org/ | | Sage | MIT License | https://github.com/lazear/sage | Sage citation: Lazear, M.R. "Sage: an open-source tool for fast proteomics searching and quantification at scale." Journal of Proteome Research 22.11 (2023): 3652–3659. Please review the licenses for any other third-party packages used. Data Privacy OxoniumBrowserX operates entirely on the local machine and does not collect, store, or transmit any data. All processing occurs locally on the user’s machine. No data is sent to external servers unless explicitly configured by the user. No Warranty Disclaimer THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT.

Citation

If you use this software in your research, please cite:

Šoić D. and Pabst M. NovoGlyco: mapping protein glycosylation in prokaryotes. bioRxiv. 2026.

Contacts

Dinko Šoić (soic@imsb.biol.ethz.ch) Martin Pabst (m.pabst@tudelft.nl)