😸 MCAAT - Metagenomic CRISPR Array Analysis Tool

• CRISPR-Cas is a bacterial immune system also famous for its use in genome editing. The diversity of known systems could be significantly increased by metagenomic data.
• Here we present the Metagenomic CRISPR Array Analysis Tool MCAAT, a highly sensitive algorithm for finding CRISPR Arrays in un-assembled metagenomic data.
• It takes advantage of the properties of CRISPR arrays that form multicycles in de Bruijn graphs.
• MCAAT's assembly-free graph-based strategy outperforms assembly-based workflows and other assembly-free methods on synthetic and real metagenomes.

---

🥳 NEWS

• Docker container available under: https://hub.docker.com/r/feeka94/mcaat
• Version 0.3 makes use of following optimization techniques:
  • Better data structures for preprocessing, phmap::flat_hash_set
  • Added compiler intrinsics to guide the hardware in the right direction
  • Reserving the capacity to prevent rehashing In depth technical details: educational resource and optimization developer notes. As a result of the above optimizations we achieved 17-25 times speedup in 1billion node graph(from 3 days to 3 hours). Considering the complexity of the graphs, this is a huge improvement.

Installation using docker

Docker Build

docker build -t mcaat .

---

Run the Tool Using Docker

Mount your working directory to access input/output files:

docker run --rm -v $(pwd):/data mcaat \
  --input_files /data/reads_R1.fastq /data/reads_R2.fastq \
  --output-folder /data/results

---

Final Image Size

The final image is based on debian:bookworm-slim and includes only:

• The mcaat binary
• Runtime libraries: libomp5, zlib1g

This keeps the image small and portable.

---

Clean Up

To remove the image:

docker rmi mcaat

Compiling the project

🔧 Build the Project

To allow ./install.sh make changes, we execute following command:

chmod +x ./install.sh

You can build the project and the working version will be saved in the build folder.

./install.sh

It is also possible to install the library by simply putting the --install flag.

./install.sh --install

To clean up you can use --clean flag.

---

Usage

./mcaat --input-files <file1> [file2] [--ram <amount>] [--threads <num>] [--output-folder <path>] [--help]

---

🧾 Command-Line Arguments

✅ Required

Argument	Description
--input_files <file1> [file2]	One or two input FASTA/FASTQ files. If one file is provided, it is treated as single-end data. If two files are provided, they are treated as paired-end reads.

⚙️ Optional

Argument	Description
--ram <amount>	Maximum RAM to use. Units: B, K, M, G. Default: 95% of system RAM Example: --ram 4G
--threads <num>	Number of threads to use. Default: total CPU cores minus 2
--output-folder <path>	Output directory for results. If not provided, a timestamped folder will be created automatically. If provided, the folder is used exactly as given.
--help, -h	Show usage information and exit

---

📁 Output Structure

The tool creates the following directory structure inside the specified output folder:

<output-folder>/
├── CRISPR_Arrays.txt         # Raw CRISPR array output

---

🧪 Example Usage

Scenario	Command
Paired-end input with custom output	./mcaat --input_files reads_R1.fastq reads_R2.fastq --ram 8G --threads 12 --output-folder results/my_run
Single-end input with default output	./mcaat --input_files reads.fastq Creates a folder like mcaat_run_2025-07-07_15-30-00/

---

Notes

• Input files must exist and be accessible.
• If RAM is set below 1 GB or above system capacity, the program will exit with an error.
• If only one input file is provided, the tool assumes single-end data.

---

⚙️ Settings file support

Create a simple key=value text file (one setting per line) and pass it with --settings /path/to/file.

The program reads values from this file unless you override them with CLI flags. If you change the file, run the program again — new values will be used.

Example of settings.txt (must include input_files):

# MUST INCLUDE
input_files=/data/sample_folder/1.fastq /data/sample_folder/2.fastq.fastq
ram=128G
threads=26
output_folder=results/run_2025-11-19
# OPTIONAL
cycle_max_length=77
cycle_min_length=27
threshold_multiplicity=20
low_abundance=true

Notes:

• input_files accepts one or two paths; entries may be separated by spaces, commas, or semicolons.
• Terminal values will override the settings.txt. For example for simplicity you can use the settings.txt file and change only -i parameter.

---

Requirements

• C++17 compiler
• RapidFuzz (for fuzzy string matching)
• Filesystem support (<filesystem>)

---

Support

If you encounter issues or have questions, feel free to open an issue or write us an email: fikrat.talibli@ibmg.uni-stuttgart.de. If you are using this software please cite this paper: https://academic.oup.com/microlife/article/doi/10.1093/femsml/uqaf016/8205558.