LysoSense CPS Analyzer

LysoSense provides a reproducible workflow for analyzing differential centrifugal sedimentation (DCS/CPS) traces collected along E. coli homogenisation campaigns. It parses instrument .dat exports, fits bi-peak Gaussian/lognormal models to quantify intact cells and inclusion bodies, and serves interactive overlays via Streamlit. The data-processing strategy is adapted from the method described in Klausser et al., 2025.

Use the Web App

• Open the hosted app: lysosense.streamlit.app
• Upload one or more CPS .dat files, inspect overlays and component fits, and download the XLSX summary.

Features

• Parse CPS/DCS .dat exports into particle_size_um vs mass_signal_ug
• Constrained bi-peak fitting (intact cells vs inclusion bodies), with single-peak fallback
• Metrics: component areas, intact fraction, lysis efficiency, mean sizes
• Interactive Plotly overlays and downloadable results table

Run Locally (for development)

python -m venv .venv
.\.venv\Scripts\activate
python -m pip install -e .
set PYTHONPATH=src
streamlit run app\streamlit_app.py

Project Layout

+- app/               # Streamlit entry point (uses Plotly for overlays)
+- data/              # Local CPS exports (ignored; keep private)
+- notebooks/         # Exploratory notebooks (e.g., cps_analyzer.ipynb)
+- src/lysosense/     # Installable package: io.py (parsing), analysis.py (fitting)
+- pyproject.toml     # Dependency + metadata definition
+- AGENTS.md          # Contributor notes
+- README.md          # You are here

Data & Privacy

• Do not commit production datasets. The data/ directory is ignored by default.
• Uploaded files in the web app are processed in-memory for analysis.

Contributing

• Follow PEP 8 (4-space indents, snake_case). Type hints for public APIs.
• Add regression tests for new parsing/fitting logic when feasible.
• Keep PRs focused (parsing vs. analysis vs. UI) and include screenshots/GIFs for UI changes.
• Suggested validation steps: streamlit run app\streamlit_app.py and relevant unit tests.

Citation

If this tool supports your work, please cite:

• Klausser et al., 2025. “Increased purity and refolding yield of bacterial inclusion bodies by recursive high pressure homogenization” Link.

Questions or feedback? Open an issue or submit a pull request.