Skip to main content

GTDB-Tk

GTDB-Tk is a software toolkit that uses machine learning to rapidly assess the quality of genome bins from metagenome-assembled genomes (MAGs). It uses the Genome Database Taxonomy (GTDB) to assign objective taxonomic classifications to bacterial and archeal genomes. This software can be used for environmental samples, isolate genomes, and single-cell genomes. Please see the source page, listed below, for more information.

Using the module​

To load GTDB-Tk in an interactive session on Palmetto:

module load biocontainers
module load gtdbtk

After loading the biocontainers module, you can see the available commands by running:

report_subtools.sh gtdbtk

To check that the database and third party programs are configured correctly for your session, using the default single thread:

gtdbtk check_install

For reference, the GTDB-Tk database is located at /datasets/GTDB-Tk/<release#>. Remember that each version of GTDB-Tk uses a specific GTDB-Tk database release.

Example​

An example workflow can be found on the GTDB-Tk website, it is also shown below:

## Step 1: Obtaining data
cd /scratch/SUSER
# Create the directory
mkdir -p gtdbtk && cd gtdbtk
# Obtain the genomes.
mkdir -p genomes
wget -q https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/003/947/435/GCF_003947435.1_ASM394743v1/GCF_003947435.1_ASM394743v1_genomic.fna.gz -O genomes/genome_a.fna.gz
wget -q https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/002/011/125/GCA_002011125.1_ASM201112v1/GCA_002011125.1_ASM201112v1_genomic.fna.gz -O genomes/genome_b.fna.gz

## Step 2: Gene calling (identify)
gtdbtk identify --genome_dir genomes --out_dir identify --extension gz --cpus $SLURM_CPUS_ON_NODE

## Step 3: Aligning genomes (align)
gtdbtk align --identify_dir identify --out_dir align --cpus $SLURM_CPUS_ON_NODE

## Step 4: Classifying genomes (classify)
gtdbtk classify --genome_dir genomes --align_dir align --out_dir classify -x gz --cpus $SLURM_CPUS_ON_NODE

## Results
# The main file output is the summary file (gtdbtk.ar53.summary.tsv). Classification of the genomes are present in the summary file.

ls classify
# classify gtdbtk.ar53.summary.tsv  gtdbtk.log  gtdbtk.warnings.log
tip

This software has multithreading features! Please try to request additional resources to expedite your analyses! Remember to set the --cpus flag accordingly!

Please see GTDB-Tk source page for more information.

Parallel Capabilities: Single core default, multithreading options available.