Skip to content

Latest commit

 

History

History
154 lines (121 loc) · 5.28 KB

setup.md

File metadata and controls

154 lines (121 loc) · 5.28 KB
Raw

Setup

Before you can run the workflows without any errors, you will need to make some preparations. Those mostly consist of downloading databases for the tools that are part of the workflow.

The second part of preparing to run a workflow consists of adjusting the params.yml file which holds all the important information for your system. To run multiple workflows (for example, first preprocessing, then assembly, lastly binning), you need to edit the params.yml file only once and then you can re-use it for all of the workflows.

The workflows here have been run on a high-performance computing cluster using the Slurm scheduler. If you are running these workflows on a different system, you might have to adjust the run.config file as well.

Table of contents

  1. Installing nextflow
  2. Databases and host reference genomes
    • [Human] reference genome
    • MetaPhlAn4 database
    • mOTUs3 database
    • CheckM database
    • VIBRANT database
    • geNomad database
  3. Adjusting the params.yml file

Installing Nextflow

On SCG, nextflow is already installed and can be loaded via module load. Please note that you need the right version of java as well for it to work. Therefore, add the following lines to your script:

module load java/18.0.2.1
module load nextflow/22.10.5

If you are running these workflows on another system, good luck! You can find a good documentation for installing Nextflow here.

Databases and host reference genomes

Host reference genomes

You will need a host reference genome on your file system so that you can remove all reads matching to the host genome from your sample, since we are interested in the bacteria, not the host genome. We map reads to the host genome with the bwa algorithm, so you will need to index the host genome before being able to use the workflow.

You can download for example the human reference genome from the UCSC Genome Browser website:

cd <your-reference-genome-location>
wget https://hgdownload.soe.ucsc.edu/goldenPath/hg38/bigZips/hg38.fa.gz
gunzip hg38.fa.gz

Now, you can index the genome with bwa. However, if you want to use the singularity container that we use for the workflow (because you might not have bwa installed on your system), you can do so like that by using the --bind command when running the singularity container:

cd <your-reference-genome-location>
singularity shell --bind ./:/mnt docker://ghcr.io/jakob-wirbel/micromamba-focal-preprocessing:latest
cd /mnt
bwa index hg38.fa

MetaPhlAn4 database

We also need to download the MetaPhlAn4 database, again using the already prepared singularity container:

cd <your-metaphlan-database-location>
singularity shell --bind ./:/mnt docker://ghcr.io/jakob-wirbel/micromamba-focal-classification:latest
cd /mnt
metaphlan --install --bowtie2db ./

mOTUs3 database

The mOTU tool needs a custom database of marker genes. Unfortunately, you cannot download the database through the tool as of now, but maybe it will be available as a feature in the future (see this issue). Instead, we can download and configure the database manually:

cd <your-motus-database-location>
wget https://zenodo.org/record/5140350/files/db_mOTU_v3.0.1.tar.gz
md5sum db_mOTU_v3.0.1.tar.gz
# expected output: f4fd09fad9b311fb4f21383f6101bfc3
# if you do not see this output, something went wrong and you need to download
# the database again!
tar -zxvf db_mOTU_v3.0.1.tar.gz

The version file in the Zenodo repository is not really up-to-date with the version numbering system of mOTUs, so you will have to adjust it manually:

cd db_mOTU
sed -i 's/2.6.0/3.0.3/g' db_mOTU_versions

CheckM database

You will also need to download the database for checkM. This is available through Zenodo as well and can be downloaded and extracted like that:

cd <your-checkm-database-location>
wget https://zenodo.org/record/7401545/files/checkm_data_2015_01_16.tar.gz
md5sum checkm_data_2015_01_16.tar.gz
# expected output: 631012fa598c43fdeb88c619ad282c4d
# if you do not see this output, something went wrong and you need to download
# the database again!
tar -zxvf checkm_data_2015_01_16.tar.gz
rm checkm_data_2015_01_16.tar.gz

VIBRANT database

tbd

geNomad database

tbd

Adjusting the params.yml file

The params.yml is THE place where information are stored about your specific workflow. For every project, it is good to have a different params.yml file, but you can use it for multiple workflows.

Make a copy of the config/params.yml file in this repository and edit it, for example using nano.

cd <your-project-location>
cp <bhattlab_workflows_nf-location>/config/params.yml ./params_projectx.yml
nano ./params_projectx.yml

Adjust the parameters in the file to those fitting your system/project. The most important parameter is the outdir parameter as it controls, where all the results will end up on your filesystem. The other parameters are mostly the location for databases or tool-specific parameters that you can adjust to control the behaviour of the tool.