METEOR Shotgun Metagenomics Snakemake Pipeline

A Snakemake pipeline for quantitative metagenomics profiling using METEOR, converted from the original Nextflow pipeline.

Overview

METEOR is a platform for quantitative metagenomics profiling of complex ecosystems. It performs:

• Species-level taxonomic profiling (Bacteria, Archaea, Eukaryotes)
• Functional analysis
• Strain-level population structure inference

Requirements

• Snakemake (≥6.0)
• Conda/Mamba
• Python ≥3.11

Installation

1. Clone or download this pipeline
2. Install Snakemake:

   conda install -c conda-forge -c bioconda snakemake

Usage

1. Prepare your data

Place paired-end FASTQ files in a directory with the naming pattern:

• *_R1*.fastq.gz and *_R2*.fastq.gz
• Or *_R1*.fastq and *_R2*.fastq

2. Configure the pipeline

Edit config.yaml:

• Set input_dir to your FASTQ directory
• Set output_dir for results
• Choose a catalogue_name or provide a custom catalogue_path
• Adjust cpus, fast mode, and memory settings as needed

3. Run the pipeline

# Dry run to check the workflow
snakemake --configfile config.yaml --dry-run

# Run with 8 cores
snakemake --configfile config.yaml --cores 8 --use-conda

# Run on a cluster (example with SLURM)
snakemake --configfile config.yaml --cores 8 --use-conda --cluster "sbatch -t 60 -c {threads}"

Pipeline Steps

1. Download/Prepare Catalogue: Downloads prebuilt microbial gene catalogue
2. FASTQ Processing: Indexes and prepares FASTQ files
3. Mapping: Maps reads against the gene catalogue
4. Profiling: Generates taxonomic and functional profiles
5. Merging: Combines profiles from all samples
6. Strain Analysis: Performs strain-level analysis (if not in fast mode)
7. Tree Building: Constructs phylogenetic trees from strain data

Output

Results are written to the specified output_dir:

• merged/: Combined taxonomic and functional profiles
• tree/: Phylogenetic trees (if strain analysis performed)
• meteor_report.html: Summary report

Configuration Options

Catalogue Options

Choose from prebuilt catalogues:

• hs_10_4_gut: Human gut microbiome
• hs_8_4_oral: Human oral microbiome
• hs_2_9_skin: Human skin microbiome
• fc_1_3_gut: Cat gut microbiome
• clf_1_0_gut: Dog gut microbiome
• mm_5_0_gut: Mouse gut microbiome
• ssc_9_3_gut: Pig gut microbiome
• And others (see config.yaml)

Processing Modes

• Normal mode: Full taxonomic + functional analysis
• Fast mode (fast: true): Taxonomic analysis only, reduced memory usage

Memory Requirements

• Normal mode: ~30GB+ RAM (depends on sample size)
• Fast mode: ~10GB RAM maximum
• Memory scales with number of reads per sample

Troubleshooting

Common Issues

1. No FASTQ files found: Check file naming pattern and input directory
2. Memory errors: Increase minimum_memory_gb or enable fast mode
3. Catalogue download fails: Check internet connection and catalogue name

Cleaning Up

Remove intermediate files:

snakemake clean --configfile config.yaml

Comparison with Original Nextflow Pipeline

This Snakemake version maintains the same functionality as nf-meteor.nf:

Feature	Nextflow	Snakemake
Catalogue download	✓	✓
FASTQ processing	✓	✓
Read mapping	✓	✓
Taxonomic profiling	✓	✓
Functional profiling	✓	✓
Profile merging	✓	✓
Strain analysis	✓	✓
Tree construction	✓	✓
Memory management	✓	✓
Fast mode	✓	✓

Citation

If you use this pipeline, please cite:

• METEOR: Ghozlane et al. "Accurate profiling of microbial communities for shotgun metagenomic sequencing with Meteor2." Microbiome (2025)
• Snakemake: Köster & Rahmann. "Snakemake—a scalable bioinformatics workflow engine." Bioinformatics (2012)