README.md

xgTaxonomy

Cross-reference of Genomic Taxonomy

Team

• Jorge M. Silva¹
• João R. Almeida¹²

1. DETI/IEETA, LASI, University of Aveiro, Aveiro, Portugal ↩
2. University of A Coruña, A Coruña, Spain ↩

How to use?

Download Project

Get xgTaxonomy project using:

git clone https://github.com/bioinformatics-ua/xgTaxonomy.git
cd xgTaxonomy/

Using Docker

To perform installation correctly, docker and docker compose must be installed in the system (see https://docs.docker.com/engine/install/ubuntu/).

Then, follow these instructions:

git clone https://github.com/bioinformatics-ua/xgTaxonomy.git
cd xgTaxonomy
docker-compose build
docker-compose up -d && docker exec -it xgTaxonomy bash && docker-compose down

Install Compressors

Give run Install Compressors for Benchmark:

bash install_compressors.sh;

Result Replication

To run the pipeline and obtain all the Reports in the folder reports, use the following commands.

Download sequences I

For obtaining random sequences for baseline test performance run:

cd src/
python3 getRandomSequences.py 

Baseline test

For baseline compression test run:

cd src/
python3 compress_baseline.py

Download sequences II

For obtaining random sequences for taxonomic classification run:

cd src/
python3 getDatabaseSequences.py 

Classifiers

Feature selection for f1-score and accuracy

cd src/
python3 classifier.py -fs -ac -b > ../results/feature_selection.txt

All columns for f1-score and accuracy

cd src/
python3 classifier.py -ac -b > ../results/f1score_accuracy_all_columns.txt

Each column individually for f1-score and accuracy

cd src/
python3 classifier.py -b > ../results/f1score_accuracy_single.txt

Classification report for each compressor

cd src/
python3 classifier.py -cr > ../results/classification_reports_single.txt

Classification f1-score and accuracy for all possible feature combinations

cd src/
python3 classifier.py -bf -b > ../results/f1score_accuracy_all_combinations.txt

Classification report for all compressors

cd src/
python3 classifier.py -cr -ac > ../results/classification_report_all_columns.txt

Classification report for all compressors

cd src/
python3 classifier.py -bf -cr > ../results/classification_report_all_combinations.txt

Cite

Please cite the following, if you use xgTaxonomy in your work:

in progress

Issues

Please let us know if there are any issues.