BGCfinder 0.0.30

Biosynthetic Gene Cluster finder with Graph Neural Network

BGCfinder : Biosynthetic Gene Cluster detection with Graph Neural Network

BGCfinder detects biosynthetic gene clusters in bacterial genomes using deep learning. BGCfinder takes a fasta file containing protein sequences and convert each of them into a graph. Graph neural network takes the input graphs to detect biosynthetic gene cluster..

• Developer : Jihun Jeung ([e-mail], [e-mail])
• Github repository : https://github.com/jihunni/BGCfinder
• PyPI project website : https://pypi.org/project/BGCfinder/

Installation requirement:

• PyTorch
• PyTorch Geometric
• Prodigal (https://github.com/hyattpd/Prodigal)

To construct the conda environment,

$ conda create --name BGCfinder  python=3.9
$ conda init bash   
$ conda activate BGCfinder   
$ conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch
$ conda install pyg -c pyg    
$ pip install BGCfinder     

To download the BGCfinder model and test files,

$ bgc-download

To find the protein-coding gene in bacterial genome (Installation of Prodigal is required for this step),

$ prodigal -f gff -i bacterial_genome_seq.fasta -a bacterial_protein_seq.fasta -o bacterial_genome_seq.gff 

To run BGCfinder with a fasta file containing amino acid sequence with CPU (recommended),

$ bgcfinder bacterial_protein_seq.fasta -o output_filename_prefix -l log_record.log -d False

To run BGCfinder with a fasta file containing amino acid sequence with GPU,

$ bgcfinder bacterial_protein_seq.fasta -o output_filename_prefix -l log_record.log -d True

The development environment of BGCfinder :

'torch==1.10.0',   
'torch-geometric==2.0.2',   
'torch-scatter==2.0.9',   
'torch-sparse==0.6.12'