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 (jihun@gm.gist.ac.kr, jeung4705@gmail.com)
• 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 --clone base
$ conda init bash
$ conda activate BGCfinder
$ conda install pytorch 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_genome.fasta -o output_filename.tsv -l log_record.log -d False

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

bgcfinder bacterial_genome.fasta -o output_filename.tsv -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'