MethArCT - Methanogenic Archaeal Culturomics Toolkit

Overview

MethArCT (Methanogenic Archaeal Culturomics Toolkit) is a comprehensive toolbox designed for metagenomic and genomic analysis of methanogenic archaea. It integrates multiple bioinformatics analysis functions to predict microbial metabolic pathways, optimal growth temperature, salinity adaptation, pH preference, antibiotic resistance, and cultivability.

Access MethArCT v0.6.0 online at (http://methardb.cn/tools/metharct-beta) for protein-based functional prediction of methanogenic archaea.

Key Features

Core Features (Required)

• Metabolic Pathway Prediction: Analysis of methane, sulfur, and nitrogen metabolic pathways (21 pathways) based on Diamond tool
• Cultivability Assessment: Evaluation of culture difficulty based on metabolic pathways

Extended Features (Optional)

• Temperature Prediction: Growth temperature range prediction (T_min, T_opt, T_max) based on AAC and dipeptide composition using the MLHuaiNanzi PLS regression model
• Salinity Prediction: Microbial salinity adaptation prediction based on amino acid composition features using SuSha ensemble learning model
• pH Prediction: Growth pH preference prediction (optimum, maximum, minimum) based on genome-wide amino acid composition features using GenomeSpot Lasso regression models
• Antibiotic Resistance Prediction: Supports prediction of Bacitracin, Tunicamycin, and Vanadate resistance.
• Genome Quality Assessment: Genome completeness and contamination estimation based on CheckM2 — requires CheckM2 tool

Quick Start

# 1. Install MethArCT
pip install .

# 2. Run comprehensive analysis
metharct comprehensive "protein.faa" -o results/

System Requirements

Basic Requirements

• Python: >= 3.9
• Required Tool: Diamond
• Operating System: Windows, Linux

Optional Tools

• CheckM2: For genome quality assessment (optional)

Quick Installation

Step 1: Get the Project

git clone https://github.com/rsj99-dev/MethArCT.git
cd MethArCT

Step 2: Create Python Environment

# Using conda (recommended)
conda env create -f environment.yml
conda activate metharct

# Or using pip with virtual environment
python -m venv metharct_env
# Windows activation:
metharct_env\Scripts\activate
# Linux/macOS activation:
source metharct_env/bin/activate

Step 3: Install Dependencies

# Install Python dependencies
pip install -r requirements.txt

# Install project as executable
pip install -e .

# Install Diamond (required)
conda install -c bioconda diamond

Windows Note: If conda install diamond fails, manually download diamond.exe from Diamond Releases and place it in the MethArCT-main root directory.

Step 4: Verify Installation

# Test core functionality
python quick_test.py

# Check command line tool
metharct --help

Optional Tools Installation

For temperature prediction or genome quality assessment features:

CheckM2 Installation (Genome Quality Assessment)

Extract "checkm2_db.zip" to the root directory of the MethArCT folder.

Usage Guide

1. Command Line Usage (Recommended)

Core Analysis (Diamond only):

# Metabolic pathway analysis — core feature
metharct diamond "protein.faa" -o results/

# Comprehensive analysis (skip optional tools)
metharct comprehensive "protein.faa" -o results/ --skip-checkm2 --skip-antibiotic

Full Analysis (with optional tools):

# Complete analysis with all features
metharct comprehensive "protein.faa" -o results/

# Run optional analyses separately
metharct susha "protein.faa" -o results/         # Salinity prediction
metharct ph "protein.faa" -o results/            # pH preference prediction
metharct antibiotic "protein.faa" -o results/    # Antibiotic resistance prediction
metharct checkm2 "genome.fasta" -o results/      # Requires CheckM2

Batch Processing (multiple genomes):

# Batch comprehensive analysis for all .faa files in a directory
metharct batch "genomes_folder/" -o results.csv

# For metagenome-assembled genomes (MAGs): show pathways with completeness >= 70%
metharct batch "genomes_folder/" -o results.csv -mags

This command processes every FASTA file (.faa, .fa, .fasta) in the given directory and generates a single summary CSV with the following columns:

Column	Description
FAA_filename	Input file name
Temperature_range	Estimated growth temperature range
Optimal_temperature	Predicted optimal growth temperature (OGT)
pH_range	Predicted pH range (min–max)
Optimal_pH	Predicted pH optimum
Salinity_range	Predicted salinity adaptation range
Substrate_metabolism	Energy metabolism pathways (methane, sulfur, nitrogen) with completeness = 100% (or ≥ 70% with -mags)
Additional_amino_acids_required	Amino acids with incomplete biosynthesis pathways
Recommended_antibiotics	Recommended antibiotics based on AAI comparison

2. Python Script Usage

from metharct import MethArCTAnalyzer

analyzer = MethArCTAnalyzer()

# Core functionality only (Diamond analysis)
results = analyzer.analyze_diamond('protein_sequences.faa')

# Antibiotic resistance prediction
results = analyzer.analyze_antibiotic('protein_sequences.faa')

# Full analysis (if optional tools are installed)
results = analyzer.analyze_comprehensive('protein_sequences.faa')

print(results)

3. Input File Requirements

• Supported Format: FASTA format (.fa, .fasta, .faa)
• File Size: Recommended not exceeding 50MB
• Character Encoding: UTF-8
• File names with special characters (e.g., parentheses, spaces) should be quoted

4. Output Description

Core Feature Output (Diamond Analysis):

• [filename]_diamond_summary.csv — Metabolic pathway summary
• [filename]_diamond_results.json — Detailed analysis results
• [filename]_*_diamond.tsv — Detailed alignment results for each pathway

Antibiotic Resistance Prediction Output:

• antibiotic_selection_results.tsv — AAI values, rule evaluation status, and recommended antibiotics

Comprehensive Analysis Report:

• [filename]_integrated_summary.csv — Integrated assessment results
• [filename]_comprehensive_analysis.json — Complete analysis data

Optional Feature Outputs:

• SuSha salinity results: [filename]_SuSha_Summary.tsv and [filename]_SuSha_Result.xlsx
• pH preference results: [filename]_pH_Summary.tsv and [filename]_pH_Details.json
• CheckM2 results in [filename]_checkm2/ directory with quality_report.tsv

Troubleshooting

Installation Issues

Command line entry point not working:

# Alternative way to run
python -m metharct.cli.main --help

# Reinstall project
pip uninstall metharct -y
pip install -e .

Diamond tool not found:

conda activate metharct
conda install -c bioconda diamond
diamond --help

Windows: Diamond not found: If conda install fails, manually download diamond.exe and place it in the MethArCT root directory:

1. Visit https://github.com/bbuchfink/diamond/releases
2. Download the latest diamond-windows.zip and extract diamond.exe
3. Place diamond.exe in the MethArCT-main root folder
4. Re-run the analysis command

pH prediction module not available:

pip install hmmlearn>=0.3.0

SuSha salinity prediction fails with module/import errors: The SuSha ensemble model requires imbalanced-learn. Make sure it is installed:

pip install imbalanced-learn>=0.10.0

MLHuaiNanzi temperature prediction not available:

pip install scikit-learn>=1.0 numpy>=1.20

LoongArch Support

Note: The LoongArch branch currently supports only the following processors: Loongson 3A/B/C/D 5000 & 6000 series, 2K2000, and 2K3000/3B6000M.

Project Structure

MethArCT/
├── metharct/                       # Main package
│   ├── cli/                        # Command line interface
│   ├── core/                       # Core analysis modules
│   │   ├── antibiotic_analyzer.py  # Antibiotic resistance prediction (AAI-based)
│   │   ├── susha/                  # SuSha salinity prediction (embedded)
│   │   ├── mlhuainanzi/              # MLHuaiNanzi temperature prediction (embedded)
│   │   └── ph_predictor/           # pH prediction engine (embedded, GenomeSpot-based)
│   │       ├── models/             # Pre-trained Lasso regression models
│   │       └── hmm/                # Signal peptide HMM model
│   └── utils/                      # Utility functions
├── data/                           # Reference databases
│   ├── databases/                  # Diamond databases
│   │   └── kangshengsu/            # Methanogen reference genomes for antibiotic prediction
│   ├── methane/                    # Methane metabolism genes
│   ├── nitrogen/                   # Nitrogen metabolism genes
│   └── sulfur/                     # Sulfur metabolism genes
├── requirements.txt                # Python dependencies
└── setup.py                        # Package setup

License

This project is licensed under the GNU General Public License v3.0 (GPL-3.0) — see the LICENSE file for details.

License history:

• Versions ≤ 0.5.5 were released under the MIT License — see LICENSE-MIT for the original terms.
• Starting from version 0.6.0, the project is relicensed under GPL-3.0 to comply with GPL-3.0 licensed dependencies (CheckM2).

Third-party components:

• CheckM2 (genome quality assessment) — GPL-3.0

Citation

If you use MethArCT in your research, please cite:

[Coming soon]