Licence: GNU General Public License v3.0 (copy provided in directory)
Author: Tom van Wijk
Contact: [email protected]

This pipeline is developed to automate the quality control, processing and analysis of large datasets of metagenome samples. Currently only Illumina paired-end data is supported.

• Linux operating system. This software is developed on Linux Ubuntu 16.04
  Experiences when using different operating systems may vary. A minimum of 8 cpu's and 64 GB of RAM are recommended
• python v2.7.x
• python libraries as listed in the import section of metagenome_pipeline.py
• FastQC v0.11.7+ https://www.bioinformatics.babraham.ac.uk/projects/fastqc/
• MultiQC v1.5+ https://multiqc.info/
• Seqkit 0.10.1+ https://bioinf.shenwei.me/seqkit/download/
• MetaPhlan2 v2.7.7+ https://bitbucket.org/biobakery/metaphlan2/src/default/
• GROOT resistome profiler https://github.com/will-rowe/groot

• Clone the metagenome_pipeline repository to the desired location on your system.
  git clone https://github.com/Papos92/metagenome_pipeline.git
• Add the location of the metagenome_pipeline repository to the PATH variable:
  export PATH=$PATH:/path/to/metagenome_pipeline
  (It is recommended to add this command to your ~/.bashrc file)
• Create path variable META_DB to the GROOT reference subdirectory:
  export META_DB=/path/to/metagenome_pipeline/groot_db
  (It is recommended to add this command to your ~/.bashrc file)
• Download the GROOT reference databases by running groot_download_db.py
  This will download the following reference databases to the groot_db directory:
  arg-annot, card, groot-core-db, groot-db and resfinder.
  
• Index the GROOT reference databases by running groot_index_db.py
  With default parameters, this will generate the index files for 2x150 bp data.
  

Start the pipeline with the following command:

dada2_amplicon_pipeline.py -i 'inputdir' -o 'outputdir' -t 'threads' -d 'database' -l 'length'

• 'inputdir': location of input directory. (required)
  Should only contain the compressed (.fastq.gz) sequence files containing the raw sequences of the forward and reverse reads. For each sample, these fastq files need to be named with an '_R1' or '_R2' tag respectively and be furthermore identical.

• 'outputdir': location of output directory. (optional)
  When not defined, a subdirectory with a timestamp will be created inside the input directory.
  

• 'threads': Number of cpu threads used.
  default = 6
  

• 'database': Reference database for GROOT resistome profiler.
  Using the recommended default option will run this analysis twice with both the resfinder and card databases and store the results in separate directories. Other options that can be used in this parameter are:
  arg-annot, card, groot-core-db, groot-db and resfinder.
  To redownload the database or download more databases when they are supported by groot in future updates, please use the groot_download_db.py script in this repository.
  

• 'length' Length of the reads in the dataset.
  Default = 150. This needs to be accurate by amargin of max. 10 bp.
  This value is referenced to use the correct database index file during the GROOT analysis.
  If you wish to use a different read length index. Please generate an index file for the databases with the groot_index_db.py script in this repository.