sccmecextractor 1.5.0

A Python toolkit for extracting SCCmec sequences from Staphylococcus whole genome sequences

SCCmecExtractor

A Python toolkit for extracting and typing SCCmec and non-mec SCC elements from Staphylococcus and Mammaliicoccus whole genome sequences. The tool identifies attachment (att) sites, extracts complete SCC elements and performs gene-level typing of mec and ccr gene complexes.

Note: extraction requires both att sites to be on the same contig as rlmH. This is a same-contig requirement with assembly fragmentation being the main source of extraction failure, particularly in non-aureus species.

Overview

SCCmecExtractor provides five CLI commands that work together to identify, extract and type SCCmec and non-mec SCC elements:

Command	Description
sccmec-pipeline	Master pipeline orchestrating all steps (recommended)
sccmec-locate-att	Locate attachment (att) sites in genomic sequences
sccmec-extract	Extract SCC elements bounded by att site pairs
sccmec-type	Type SCC elements by mec complex class (A-E), ccr complex type (1-22) and SCCmec type (I-XV)
sccmec-report	Merge extraction and typing results into a unified report

Key Capabilities

• FASTA-only mode — no GFF annotation required; rlmH detected via BLAST against a 70-species reference database
• Non-mec SCC detection — extracts SCC elements that carry ccr genes but lack mec genes
• Composite element detection — identifies tandem/nested SCC elements with multiple att site pairs
• SCCmec typing — assigns mec complex class (A-E), ccr complex type (1-22 per IWG-SCC) and SCCmec type (I-XV) via BLAST-based gene detection, including IS element proximity and orientation analysis
• Cross-genus support — validated on Staphylococcus (64 species) and Mammaliicoccus (6 species)

Table of Contents

• Installation
  • Using Conda/Mamba
  • Using pip
  • Using Docker
  • Using Singularity
• Requirements
• Usage
  • FASTA-only Mode (Recommended)
  • GFF Mode
  • Command Reference
• Complete Workflow Examples
• How It Works
• Output Format
• Troubleshooting
• Citation
• License
• Contributing
• Contact

Installation

Using Conda/Mamba (Recommended)

# Create a new environment
conda create -n sccmecextractor python=3.11
conda activate sccmecextractor

# Install dependencies
conda install -c conda-forge -c bioconda biopython blast

# Install Bakta only if you need GFF-based workflow
conda install -c conda-forge -c bioconda bakta

# Install SCCmecExtractor
pip install sccmecextractor

# Test that commands are available
sccmec-pipeline --help
sccmec-locate-att --help
sccmec-extract --help
sccmec-type --help
sccmec-report --help

If commands do not run, make sure the environment's bin/ directory is in your PATH:

export PATH="$CONDA_PREFIX/bin:$PATH"

Using pip

Note: installation with pip does not provide BLAST+ or Bakta. BLAST+ is required for ccr gene checks, typing and FASTA-only mode.

# Install SCCmecExtractor
pip install sccmecextractor

# Test that commands are available
sccmec-pipeline --help

Using Docker

Docker provides a containerised environment with all dependencies pre-installed, including BLAST+ and Bakta.

# Pull the pre-built image
docker pull alisonmacfadyen/sccmecextractor:latest

# Or build from source
git clone https://github.com/AlisonMacFadyen/SCCmecExtractor.git
cd SCCmecExtractor
docker build -t sccmecextractor:latest -f containers/Dockerfile .

Quick Start with Docker:

# FASTA-only mode (no Bakta database needed)
docker run --rm \
  -v $PWD:/work \
  alisonmacfadyen/sccmecextractor:latest \
  sccmec-pipeline --fna-dir genomes/ -o results/

# With GFF annotation
docker run --rm \
  -v $PWD:/work \
  alisonmacfadyen/sccmecextractor:latest \
  sccmec-pipeline -f genome.fna -g genome.gff3 -o results/

See CONTAINER_GUIDE.md for detailed Docker usage instructions.

Using Singularity

Singularity/Apptainer is ideal for HPC environments where Docker is not available.

# Pull from Docker Hub
singularity pull docker://alisonmacfadyen/sccmecextractor:latest

# Or build from definition file
singularity build sccmecextractor.sif containers/sccmecextractor.def

Quick Start with Singularity:

# FASTA-only mode
singularity exec \
  --bind $PWD:/work \
  sccmecextractor.sif \
  sccmec-pipeline --fna-dir genomes/ -o results/

# With GFF annotation
singularity exec \
  --bind $PWD:/work \
  sccmecextractor.sif \
  sccmec-pipeline -f genome.fna -g genome.gff3 -o results/

See CONTAINER_GUIDE.md for detailed Singularity usage instructions.

Requirements

Dependencies

• Python 3.9+
• Biopython
• BLAST+ (bundled in containers)
• Bakta (optional, only needed for GFF-based workflow) and automatically included in containers

Input Files

• Genome sequence: .fasta or .fna file containing the assembled genome

When using GFF mode:

• Gene annotations: .gff3 file with gene annotations (we recommend Bakta for annotation)

When using FASTA-only mode (--blast-rlmh): No annotation file is needed rlmH is detected via BLAST against a bundled 70-species reference database

Bakta Database

Only required if using Bakta for GFF annotation:

# Light database (faster, smaller)
bakta_db download --output bakta_db --type light

# Full database
bakta_db download --output bakta_db

Usage

FASTA-only Mode (Recommended)

The simplest way to run SCCmecExtractor. No GFF annotation or Bakta database is required — rlmH is detected via BLAST. If using a novel species, recommended to supply rlmH gene reference.

# Single genome
sccmec-pipeline -f genome.fna -o results/

# Multiple genomes
sccmec-pipeline -f genome1.fna genome2.fna genome3.fna -o results/

# Directory of genomes
sccmec-pipeline --fna-dir genomes/ -o results/

# With composite element detection and multithreading
sccmec-pipeline --fna-dir genomes/ --composite -o results/ -t 4

GFF Mode

If you have pre-computed GFF3 annotations (e.g. from Bakta), you can provide them. GFF files are matched to FASTA files by stem name.

# Single genome with GFF
sccmec-pipeline -f genome.fna -g genome.gff3 -o results/

# Directories of genomes and GFFs
sccmec-pipeline --fna-dir genomes/ --gff-dir annotations/ -o results/

Command Reference

sccmec-pipeline

Master pipeline that runs all steps: att site location, extraction, typing and report generation.

sccmec-pipeline [-h] (-f FNA [FNA ...] | --fna-dir FNA_DIR)
                [-g GFF [GFF ...] | --gff-dir GFF_DIR] [--blast-rlmh]
                [--rlmh-ref RLMH_REF] [--composite] -o OUTDIR [-t THREADS]
                [--mec-ref MEC_REF] [--ccr-ref CCR_REF]

Argument	Description
-f, --fna	One or more FASTA/FNA genome files
--fna-dir	Directory of FASTA/FNA genome files
-g, --gff	One or more GFF3 annotation files (matched by stem name)
--gff-dir	Directory of GFF3 files (matched by stem name)
--blast-rlmh	Use BLAST for rlmH detection (auto-enabled when no GFF provided)
--rlmh-ref	Custom rlmH reference FASTA for BLAST detection
--composite	Extract to outermost boundary for composite elements
-o, --outdir	Output directory for all results
-t, --threads	Number of parallel threads (default: 1)
--mec-ref	Custom mec gene reference FASTA for gene content detection (see sccmec-type)
--ccr-ref	Custom ccr gene reference FASTA for gene content detection (see sccmec-type)

sccmec-locate-att

Identifies attachment sites in genomic sequences.

sccmec-locate-att [-h] -f FNA [-g GFF] -o OUTFILE [--blast-rlmh] [--rlmh-ref RLMH_REF]

Argument	Description
-f, --fna	Input genome file (.fasta or .fna)
-g, --gff	Gene annotation file (.gff3 format, optional)
-o, --outfile	Output TSV file containing att site locations
--blast-rlmh	Use BLAST for rlmH detection (auto-enabled when no GFF provided)
--rlmh-ref	Custom rlmH reference FASTA

sccmec-extract

Extracts SCC elements bounded by att site pairs.

sccmec-extract [-h] -f FNA [-g GFF] -a ATT -s SCCMEC [--composite] [-r REPORT]
               [--blast-rlmh] [--rlmh-ref RLMH_REF]

Argument	Description
-f, --fna	Input genome file (.fasta or .fna)
-g, --gff	Gene annotation file (.gff3 format, optional)
-a, --att	TSV file from sccmec-locate-att with att site locations
-s, --sccmec	Output directory for extracted SCC sequences
--composite	Extract to outermost boundary for composite elements
-r, --report	Output TSV file for extraction report (appends for batch)
--blast-rlmh	Use BLAST for rlmH detection (auto-enabled when no GFF provided)
--rlmh-ref	Custom rlmH reference FASTA

sccmec-type

Types SCC elements (or whole genomes) by mec complex class (A-E), ccr complex type (1-22) and SCCmec type (I-XV). Classification is based on BLAST detection of mec genes, ccr allotypes and mec complex structural genes (IS431, IS1272, mecI, mecR1), with IS element proximity and orientation analysis for mec class determination.

sccmec-type [-h] -f FASTA [FASTA ...] -o OUTFILE [--mec-ref MEC_REF] [--ccr-ref CCR_REF]

Argument	Description
-f, --fasta	Input FASTA file(s) or directory of extracted SCC elements
-o, --outfile	Output TSV file for typing results
--mec-ref	Custom mec gene reference FASTA for gene content detection. When provided, mec genes are detected using this reference; mec complex class and SCCmec type will not be assigned. ccr typing still uses the bundled reference
--ccr-ref	Custom ccr gene reference FASTA for gene content detection. When provided, ccr genes are detected using this reference; ccr complex type and SCCmec type will not be assigned. mec typing still uses the bundled reference

sccmec-report

Merges extraction and typing results into a unified report.

sccmec-report [-h] -e EXTRACTION_REPORT -t TYPING_RESULTS -o OUTFILE

Argument	Description
-e, --extraction-report	TSV from sccmec-extract --report
-t, --typing-results	TSV from sccmec-type
-o, --outfile	Output unified report TSV

Complete Workflow Examples

FASTA-only Mode (Recommended)

# One-liner for a directory of genomes
sccmec-pipeline --fna-dir genomes/ -o results/ -t 4

GFF Mode with Bakta Annotation

# 1. Annotate your genome with Bakta
bakta --db bakta_db genome.fna --output bakta_output

# 2. Run the pipeline with GFF
sccmec-pipeline -f genome.fna -g bakta_output/genome.gff3 -o results/

Step-by-Step (Individual Commands)

# 1. Locate att sites
sccmec-locate-att -f genome.fna -o att_sites/att_sites.tsv

# 2. Extract SCC elements
sccmec-extract -f genome.fna -a att_sites/att_sites.tsv -s sccmec/ -r extraction_report.tsv

# 3. Type extracted elements and whole genome
sccmec-type -f sccmec/ genome.fna -o typing_results.tsv

# 4. Generate unified report
sccmec-report -e extraction_report.tsv -t typing_results.tsv -o unified_report.tsv

Docker

# FASTA-only mode (simplest)
docker run --rm \
  -v $PWD:/work \
  alisonmacfadyen/sccmecextractor:latest \
  sccmec-pipeline --fna-dir genomes/ -o results/

# With Bakta annotation
docker run --rm \
  -v $PWD:/work \
  -v ~/bakta_db:/data/bakta_db \
  alisonmacfadyen/sccmecextractor:latest \
  bash -c "
    bakta --db /data/bakta_db genome.fna --output bakta_output --prefix genome && \
    sccmec-pipeline -f genome.fna -g bakta_output/genome.gff3 -o results/
  "

Singularity

# FASTA-only mode (simplest)
singularity exec \
  --bind $PWD:/work \
  sccmecextractor.sif \
  sccmec-pipeline --fna-dir genomes/ -o results/

# With Bakta annotation
singularity exec \
  --bind $PWD:/work \
  --bind ~/bakta_db:/data/bakta_db \
  sccmecextractor.sif \
  bash -c "
    bakta --db /data/bakta_db genome.fna --output bakta_output --prefix genome && \
    sccmec-pipeline -f genome.fna -g bakta_output/genome.gff3 -o results/
  "

How It Works

Attachment Site Detection

The tool searches for 24 DNA motif patterns (8 attR/cattR + 16 attL/cattL) representing the attachment sites that flank SCC elements. These patterns use regex with degeneracy to account for sequence variation across species. attR sites are anchored within the rlmH gene, which serves as the chromosomal integration site.

• attR/cattR: Right attachment sites (within rlmH), including CcrC-associated attR2/cattR2 and attR3/cattR3 variants
• attL/cattL: Left attachment sites, including multiple bridge-length variants (attL2-attL8) covering diverse ccr-mediated recombination products

Extraction Logic

1. rlmH Detection: Locates the rlmH gene via GFF annotation or BLAST against a 70-species reference database
2. Site Validation: attR sites must fall within or near rlmH (within 100 bp)
3. Pair Selection: Identifies the closest attR-attL pair on the same contig
4. ccr Validation: Verifies ccr genes are present between the att sites. Elements without ccr are classified as no_ccr_element (not extracted)
5. Size Filtering: Rejects artefacts < 1,000 bp (pattern overlaps) and spurious matches > 200,000 bp
6. Composite Detection: Identifies tandem/nested elements with multiple att site pairs (with --composite)
7. Origin-Spanning Detection: Flags elements where chromosome linearisation splits the SCC across contig ends
8. Fallback Extraction: When standard extraction fails, but attL is identified, fallback extraction is utilised using the location of rlmH as a proxy for attR
9. Strand Awareness: Automatically handles reverse complement extraction when necessary

SCCmec Typing

sccmec-type performs three-tier classification via BLAST:

1. mec complex class (A, B, C1, C2, D, E) — determined by mecA/mecC allotype, presence/absence of mecI, IS1272, IS431 and mecR1, and IS431 orientation. IS elements are only counted when within 10 kbp of mecA on the same contig
2. ccr complex type (1-22) — determined by ccrA/ccrB pair and ccrC allotype combinations, incorporating IWG-SCC types 1-9 and Huang et al. 2024 types 10-22
3. SCCmec type (I-XV) — assigned by combining mec class and ccr complex type. Unrecognised combinations are reported as novel_combination

Output Format

Pipeline Output Directory Structure

results/
├── ambiguous_att_sites.tsv    # Collates extraction-limited genomes that may be of interest
├── att_sites/                 # att site locations (one TSV per genome)
│   └── *.tsv
├── sccmec/                    # Extracted SCC element FASTAs
│   └── *_SCCmec.fasta
├── sccmec_unified_report.tsv  # Full diagnostic report (extraction + typing merged)
└── sccmec_summary.tsv         # Concise one-line-per-genome summary

Unified Report

The unified report (sccmec_unified_report.tsv) includes per-genome columns for extraction status, att site coordinates, element size, mec/ccr gene content, allotype classifications, mec complex class, ccr complex type, SCCmec type, mec context (location of mec genes relative to the extracted element) and typing source (sccmec or wgs).

A concise summary report (sccmec_summary.tsv) is also generated for quick screening of large datasets.

Troubleshooting

Common Issues

No att sites found:

May be genuine due to non-matching patterns, however:

• Verify that the input FASTA file is properly formatted
• If using GFF mode, ensure the GFF3 file corresponds to the same genome assembly

No rlmH gene found:

• In FASTA-only mode, rlmH is detected via BLAST. If your species has a highly divergent rlmH, provide a custom reference with --rlmh-ref
• In GFF mode, verify that gene annotation was performed correctly and rlmH is annotated as such
  • The tool expects either the gene annotated as rlmH or the product is named "Ribosomal RNA large subunit methyltransferase H" (case-insenstive)
• rlmH is a conserved housekeeping gene — its absence may indicate an assembly issue

Missing attR-attL pairs (cross_contig):

• Caused by assembly fragmentation, the att sites are on different contigs
• Check the att site TSV output to see which sites were detected and on which contigs
• Long-read sequencing or hybrid assembly can improve extraction rates

Missing attR-attL pairs (right_only):

• Missing att pairs will most commonly occur due to a missing attL, therefore SCC elements cannot be extracted.
• This is a current limitation of the tool

BLAST+ not found:

• BLAST+ is required for FASTA-only mode (--blast-rlmh), typing and ccr location checks
• Install via conda: conda install -c bioconda blast
• BLAST+ is pre-installed in Docker and Singularity containers

Container-specific issues:

• See CONTAINER_GUIDE.md for troubleshooting Docker and Singularity problems

Warning Messages

The tools provide informative warning messages to help diagnose issues:

• Missing gene annotations
• Incomplete att site pairs
• File processing errors

Citation

If you use SCCmecExtractor in your research, please cite the bioRxiv Preprint::

MacFadyen, A.C. SCCmecExtractor: A tool for extracting Staphylococcal Cassette Chromosome elements from Whole Genome Sequences. bioRxiv 2026.03.31.715619; doi: https://doi.org/10.64898/2026.03.31.715619

License

MIT License

Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.

Contact

Email: [e-mail]

Acknowledgments

• Bakta for bacterial genome annotation
• The Biopython project for sequence manipulation tools
• BLAST+ for sequence comparison/alignment