Image Cytometry and Kinetic Modelling Reveal How Aged Leaf Biomass Dose Regulates Microbial Physiology and PAH Degradation

Imaging Cytometry and Kinetic Modelling Repository

This repository contains the complete processed dataset, trained machine learning models, and reproducible analysis framework supporting the study:

“Image Cytometry and Kinetic Modelling Reveal How Aged Leaf Biomass Dose Regulates Microbial Physiology and PAH Degradation.”

Study Overview

This dataset integrates:

• First-order PAH degradation kinetics

• Imaging flow cytometry–derived single-cell morphological features

• Supervised machine learning classification of microbial physiological states

• Probability calibration and deployment thresholds

The repository enables full reproduction of:

• Viable vs debris classification

• Single vs aggregate discrimination

• Fluorescence-defined state classification

• Morphology–kinetic correlation analyses

Repository Contents

• ✔ Processed feature dataset (~4.5 × 10⁵ segmented cellular objects)

• ✔ Trained models (.joblib)

• ✔ Threshold calibration file

• ✔ Inference script (predict.py)

• ✔ Reliability calibration outputs

• ✔ Version-locked dependency file

Computational Environment

• Python 3.11

• scikit-learn 1.4.2

• NumPy 1.26

• Pandas 2.2

• Matplotlib 3.8

Purpose

This repository provides a reproducible framework linking amendment dosage, microbial physiological integrity, and PAH degradation kinetics through morphology-derived phenotyping.