Overview
The StrepSuis-AMRPat module provides automated detection and analysis of antimicrobial resistance patterns in Streptococcus suis. It identifies multi-drug resistant (MDR) strains and analyzes co-resistance patterns across antibiotic classes using advanced statistical methods and network analysis.
Key Features
- MDR Classification: Customizable threshold for defining MDR strains (default: resistance to ≥3 classes)
- Bootstrap Prevalence: Robust prevalence estimation with confidence intervals
- Association Rules: Mining patterns in resistance gene co-occurrence
- Hybrid Network: Combined phenotype-genotype resistance network
- Community Detection: Identifying resistance gene clusters
When to Use
Use this tool when you want to:
- Identify and characterize MDR strains in your collection
- Discover co-resistance patterns across antibiotic classes
- Find associations between resistance genes
- Understand resistance network structure
Input Files Required
resistance_data.csv - Combined resistance data (MIC + AMR genes)
Note: Can also merge separate MIC.csv and AMR_genes.csv files
Output Files
- HTML Report: Interactive MDR classification and network visualizations
- Excel Workbook: MDR strains list, association rules, network metrics
- PNG Charts: Network diagrams, prevalence plots, heatmaps
Statistical Methods
- Bootstrap resampling for prevalence estimation
- Chi-square and Fisher exact tests for associations
- Apriori algorithm for association rule mining
- Louvain algorithm for community detection
- Benjamini-Hochberg FDR correction
Quick Start
Option 1: Google Colab (Recommended)
- Click "Run in Colab" button
- Execute all cells
- Upload resistance data file
- Set MDR threshold (default: 3)
- Download comprehensive results
Run in Colab
Option 2: Local Installation
# Clone and setup
git clone https://github.com/MK-vet/MKrep.git
cd MKrep
pip install -r requirements.txt
# Run analysis
python MDR_2025_04_15.py
Option 3: CLI
mkrep-mdr --data-dir ./data --output ./results --mdr-threshold 3