h3ameta

H3ABionNet Metagenomics Workflows

Note: other workshop materials can be found in our Google Drive folder

Running the model workflow

1. Set up conda, nextflow, clone the Git repository.

Note: this requires Singularity to be set up on your system or cluster.

cd ~
mkdir -p ~/local/bin
export PATH="$PATH:~/local/bin"

wget -qO- https://get.nextflow.io | bash
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
cd
git clone https://github.com/h3abionet/h3ameta.git

2. Running the workflow

cd ~
mkdir test_run; cd test_run
nextflow h3ameta/examples/taxonomic_classification/taxonomic_classification.nf  --tax_level S -resume --in h3ameta/examples/test_data/*.fq \
--dataset_table h3ameta/examples/test_data/datasets.tsv --db /path/to/kraken_and_bracken_db

Docker images

We're assuming you're using singularity -- if using Docker it'll be a little simpler, so it's left as an exercise for the reader. Of course, if you're using Nextflow this will generally be taken care of by the appropriate config file and should be transparent.

kraken2

Download the latest image

singularity pull docker://quay.io/h3abionet_org/kraken2

This will create an image kraken2.img which contains the kraken2 suite plus auxiliary programs like dustmasker

Note that we do not have any databases inside the image to keep the image small. You need to download and build the databases. Here's an example: Assume that you have a directory /local/kraken and you're going to bulild the database inside that

singularity exec -B /local/kraken/:/mnt kraken2.simg kraken2-build --standard --threads 8 --db /mnt/krakdb

This binds the directory /local/kraken on the host to the /mnt directory in the singularity image. The directory /mnt is passed to the kraken2-build program to use for the data and the database will be called krakdb.

Funding

We acknowledge support of the NIH (Grant U24HG006941)