Nmdp-Flow

Consensus assembly, variant calling, and allele interpretation workflow.

This version of the pipeline was used to assess the impact of Docker containers on the performance of genomic tools. See https://peerj.com/preprints/1171/ and subsequent manuscript peer review https://peerj.com/manuscripts/5515/

To replicate the pipeline execution see the following section.

Quick start

Make sure you have the required dependencies listed in the last section.

1. Pull the required Docker image:

   $ docker pull nextflow/nmdp-flow:peerj5515

2. Download the benchmark dataset:

   $ aws s3 sync s3://cbcrg-eu/nmdpflow-data data

3. Launch the pipeline execution with the benchmark dataset

   $ nextflow run nmdp-flow -revision peerj5515 -with-docker -bg > log.txt

The pipeline will run as a background process and the output redirected to the log.txt file.

It will also create a file named trace.csv containing the run time information for each executed task.

Note: the execution with Docker will produce some intermediate results with root ownership. You will need sudo/root permissions to delete them.

Dependencies

• Java VM 7 or later
• Docker 1.0 or later
• Nextflow 0.15.0 or later
• AWS CLI tools

For the execution of the pipeline without using the Docker engine follow the installation steps in the included Dockerfile.