ONN4ARG is an Ontology-aware Neural Network model for Antibiotic Resistance Gene (ARG) annotation predictions. It employs a novel ontology-aware layer to encourage annotation predictions satisfying the ontology rules (i.e., the ontology tree structure). It requires the Diamond and the HHblits alignment tools to run. Our source codes are available on GitHub, and our pre-built ARG database and our pre-trained model can be downloaded from Zenodo.
./predict.sh FASTA_fileprefix
The program will take "FASTA_fileprefix.fasta" as input and store the predicted annotations in "FASTA_fileprefix.out".