PFASProphet 0.1.0

ML-based PFAS prediction from mass spectrometry data

PFASProphet

A tool for giving a score on the likelihood of PFAS (Per- and Polyfluoroalkyl Substances) from mass spectrometry data using machine learning.

Features

• Obtain a PFAS scores from precursor masses and fragment ions.
• Support for CSV files or direct list inputs.
• can be run from python directly or Command-line interface (CLI).
• Handles ionised vs. neutral masses (negative ESI mode).
• See examples for usage.

Installation

1. Clone the repository: git clone https://github.com/yourusername/PFASProphet.git
2. Navigate to the directory: cd PFASProphet
3. Create a virtual environment: python -m venv .venv
4. Activate it: .venv\Scripts\activate (Windows) or source .venv/bin/activate (Linux/Mac)
5. Install: pip install -e .

Usage

CLI Examples

• Predict from lists: pfasprophet --mass "[248.9461]" --fragments "[[63.9624]]"
• Predict from CSV: pfasprophet --file data.csv
• Help: pfasprophet or pfasprophet --help

Python API

from PFASProphet import PFASProphet

prophet = PFASProphet()
result = prophet.predict(mass=[248.9461], fragments=[[63.9624]])
print(result)