GraphK-LR : Long-reads metagenomic bin refiner

Dependencies

A possible conda environment to work

For a system with CUDA:

conda env create -f environment.yml

For a system with only CPU:

conda env create -f env_cpu.yml

To download GraphK-LR, you have to clone the GraphK-LR repository to your machine.

git clone https://github.com/NethmiRanasinghe/GraphK-LR.git

cd GraphK-LR

sh build.sh

Download dataset from here;

A result file obtained from running OBLR (link) can be found inside testData folder. First, convert this file to the format preferred by our tool.

python graphk/support/results_parser.py -i testData/OBLR_result.npz -r reads.fasta -o testData -t oblr

This will create the file initial_bins.npy.

Run GraphK-LR;

python main.py -r <path to fastq/fasta file> -i <path to initial_bins.npy> -o <path to output folder>

Use the -h argument to list all the available commands.