No upload. No limits. 100% local.
Developed by PRISM U1192 Laboratory — INSERM / CHU de Lille / Université de Lille
⬇️ Download Latest Release · 🏠 Profiler Homepage · 🌐 Web Version · 📦 Test Datasets · 📄 Paper
Profiler Desktop v1.2 is the standalone, fully offline version of Profiler — an interactive platform for multi-omics data analysis. It runs entirely on your local machine: no internet connection, no data upload, no account required.
| Web version | Desktop v1.2 | |
|---|---|---|
| Installation | None | Automated (one double-click) |
| Internet required | Yes | No |
| Upload size limit | Restricted | Unlimited |
| Data privacy | In-session | 100% local |
| All features | ✓ | ✓ |
| HTML report export | ✓ | ✓ |
| Raw data conversion | ✓ | ✓ (optional) |
- GSEA — Gene Set Enrichment Analysis from any output (volcano, heatmap, Venn/UpSet), joining ORA across 100+ databases
- Regression modeling — ML and MLP for continuous targets; R², RMSE, residual plots, cross-validation
- Longitudinal analysis — mixed-effects models, trajectory visualisation, repeated-measures statistics
- HTML report generator — one-click self-contained export of all session plots, tables, and metrics
- Clinical metadata support — any column ending with
_metaused as clinical covariate or classification target - Extended format support — Spectronaut, FragPipe, DESeq2/edgeR, Salmon, kallisto, MetaboAnalyst, XCMS, MZmine and more
- Refactored architecture — clean
app/package structure, robust installer,protobufconflict resolved
💡 Administrator rights are not required.
⚠️ Important: During the installation of Miniconda or Anaconda, make sure to check the option "Add to my PATH environment variable" (or similar).
Otherwise, thecondacommand will not be recognized in your terminal.
Download the latest ZIP from either:
Extract it anywhere on your machine (e.g. C:\Users\YourName\Desktop\Profiler_Desktop1.2).
⚠️ Do not run anything before extracting the ZIP.
Open the extracted folder and double-click install_profiler.bat.
A terminal window will open and run automatically. Installation takes a few minutes.
What the installer does:
- Verifies Conda is available
- Accepts Conda Terms of Service automatically (conda ≥ 24.x)
- Creates a
profilerconda environment (Python 3.8.20) - Installs all Python dependencies from
requirements.txt - Attempts to install ProteoWizard / msconvert (optional — for RAW file conversion only)
- Creates a "Profiler Desktop" shortcut on your Desktop
If ProteoWizard cannot be downloaded automatically, Profiler still works fully — only RAW file conversion is affected. Manual install instructions are displayed.
Double-click the "Profiler Desktop" shortcut on your Desktop, or run run_profiler.bat directly.
Profiler will open in your default browser. No terminal, no commands, fully local.
⏱️ First launch may take 30–60 seconds while Streamlit initialises.
💡 Administrator rights are not required.
⚠️ Important: During the installation of Miniconda or Anaconda, make sure to check the option "Add to my PATH environment variable" (or similar).
Otherwise, thecondacommand will not be recognized in your terminal.
# 1. Create environment
conda create -n profiler python=3.8.20
conda activate profiler
# 2. Install dependencies
cd /path/to/Profiler_Desktop1.2
pip install -r requirements.txt
pip install tensorflow==2.10.1 --no-deps
# 3. Set protobuf compatibility flag (required for tensorflow 2.10.1)
export PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python # Linux/macOS
# or on Windows:
# set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python
# 4. Launch
python app/main.pyProfiler_Desktop1.2/
│
├── install_profiler.bat ← Windows installer (double-click to install)
├── setup_environment.ps1 ← PowerShell setup script (called by installer)
├── run_profiler.bat ← Windows launcher (double-click to run)
├── requirements.txt ← Python dependencies
│
└── app/
├── main.py ← Entry point
├── gui/
│ └── Profiler_Desktop_Gui.py ← Main Streamlit interface
├── core/
│ ├── profiler_preprocessing.py ← Normalisation & binning
│ ├── profiler_DL.py ← MLP / CNN / RNN
│ └── profiler_training.py ← Classical ML
├── analysis/
│ ├── profiler_features_importance.py
│ ├── profiler_genes_enrichment.py
│ ├── profiler_survival.py
│ ├── profiler_unsupervised.py
│ ├── profiler_visualization.py
│ ├── profiler_normality.py
│ ├── profiler_longitudinal.py
│ ├── profiler_rt.py
│ └── profiler_sampling.py
├── data/
│ ├── profiler_data_loading.py
│ ├── profiler_data_exploration.py
│ ├── profiler_conversion.py
│ └── profiler_structured_data_file.py
├── utils/
│ ├── profiler_imports.py
│ ├── session_store.py
│ └── reset_data_session.py
└── assets/
├── profiler_logo.png
└── profiler_icons.ttf
Located in Additional_tools/MSI2Profiler/ — a companion tool for Mass Spectrometry Imaging (MSI) preprocessing.
Supports .imzML files from MALDI-MSI and DESI-MSI. Outputs a Profiler-ready CSV matrix.
MSI2Profiler is available as a Windows executable — no installation, no Python required.
👉 Download MSI2Profiler for Windows (.exe)
Just double-click and run. No Conda, no terminal, no setup.
Also available on the Profiler homepage.
pip install pandas numpy plotly pyimzml
python MSI2profiler.pyFull documentation: Additional_tools/MSI2Profiler/
| Component | Minimum | Recommended |
|---|---|---|
| OS | Windows 10 64-bit | Windows 11 |
| RAM | 16 GB | 32 GB |
| CPU | 4 cores | 8 cores |
| Storage | 3 GB | 10 GB+ |
If you use Profiler or Profiler Desktop in your research, please cite:
Zirem, Y., Ledoux, L., Fournier, I., & Salzet, M. Profiler: an open web platform for multi-omics analysis. Bioinformatics, Oxford University Press, 2025. DOI: 10.1093/bioinformatics/btaf644 PMID: 41324558
Example datasets available at: 👉 https://github.com/yanisZirem/Profiler_v1_requests_datatests
Includes: MaxQuant / DIA-NN outputs, Bruker & Waters RAW files, multi-omics tabular data, survival data, longitudinal data, and peer-review paper datasets.
Profiler Desktop is proprietary software registered with the Agence pour la Protection des Programmes (APP).
IDDN Certificate: IDDN.FR.001.300044.000.S6.C7.2025.0009.3123010
All rights reserved. See License & Intellectual Property.txt for full terms.
For licensing or collaboration: yanis.zirem@univ-lille.fr
Yanis Zirem — PhD Candidate, PRISM U1192 📧 yanis.zirem@univ-lille.fr
Supervised by Prof. Michel Salzet and Prof. Isabelle Fournier PRISM U1192 — INSERM / CHU de Lille / Université de Lille