/Archaea2

Classification and identification of the Archaeas.

Primary LanguagePythonMIT LicenseMIT

Archaea2


Using Docker

git clone git@github.com:jorgeMFS/Archaea2.git
cd Archaea2
docker-compose build
docker-compose up -d && docker exec -it archaea2 bash && docker-compose down

Install Tools

Get Archaea project using:

chmod +x run.sh
bash run.sh 

Scripts

Dataset Download:

cd scripts || exit
bash download_dataset.sh

Preprocess Dataset:

cd scripts || exit
bash prepare_and_classify_dataset.sh

Classification:

cd "../python_src/" || exit
python3 create_classification_dataset.py
python3 classification.py

Project structure

  • /scripts: bash scripts
  • /python_src: python3 scripts
  • /reports: extrated features from pre-processing dataset
  • /taxonomic_info: sample taxonomic description and mapping to taxonomic group
  • /data: numpy data to perform classification
  • /xlslist: classification results.

Requirements

  • Ubunto 18.0 or higher
  • Python3.6
  • Anaconda
  • or docker and docker-compose.

Cite

Please cite this work as:

Silva, Jorge Miguel, et al. "Feature-Based Classification of Archaeal Sequences Using Compression-Based Methods." Iberian Conference on Pattern Recognition and Image Analysis. Springer, Cham, 2022.

BibTeX

@inproceedings{silva2022feature,
  title={Feature-Based Classification of Archaeal Sequences Using Compression-Based Methods},
  author={Silva, Jorge Miguel and Pratas, Diogo and Caetano, T{\^a}nia and Matos, S{\'e}rgio},
  booktitle={Iberian Conference on Pattern Recognition and Image Analysis},
  pages={309--320},
  year={2022},
  organization={Springer}
}