16SProcessing

A Nextflow workflow to run vsearch for short read 16S data. The workflow includes the following steps:

1. Trimming (cutadapt) and QC (fastQC)
2. Merging and filtering reads (vsearch)
3. Clustering and chimera removal (vsearch and swarm)
4. OTU classification (vsearch sintax)

Notes:

• The workflow is configured to work with docker or singularity. The singularity profile works with SLURM by default, an sbatch job can be submitted with the available example script.
• Fastq files must be named *_L001_R{1,2}_001.fastq.gz

Setup:

• Install Nextflow
• Install Singularity or Docker
• Move to 'data' folder in main directory and download and unzip:
  • https://www.drive5.com/sintax/rdp_16s_v18.fa.gz
  • https://mothur.s3.us-east-2.amazonaws.com/wiki/silva.gold.bacteria.zip
  • Run unzip -p silva.gold.bacteria.zip | sed -e "s/[.-]//g" > gold.fasta
  • Run gunzip rdp_16s_v18.fa.gz
• Edit the file nextflow.config, edit runOptions to include the actual path of the data file in your environment

To run:

nextflow run 16SProcessing.nf --in_dir directory/with/fastq/files -profile (docker OR singularity)

Workflow can be tested with:

nextflow run 16SProcessing.nf --in_dir test_16S_reads -profile (docker OR singularity)