Bisulfite Nextflow Pipeline

Step 1: Clone pipeline into your project directory that contains the fastq files.

git clone https://github.com/Apompetti-Cori/Bisulfite_Nextflow.git

You will now have a folder in your directory called Bisulfite_Nextflow.

Step 2: Create csv file explaining the samples

• Use sample_table_template.csv as a guide for your sample table.
• First column should be sample containing sample ID's
• Consequent columns should be r1_L1, r1_L2, r1_L3, r1_L4, r2_L1, r2_L2, r2_L3, r2_L4
  • Example: a single end fastq should have a sample ID and a file name inhabiting the r1_L1 column
    • If it is multilane, it should have the other lanes inhabiting the other r1_L* columns
  • Example: a paired end fastq should have a sample ID and file names inhabiting the r1_L1 and r2_L1 columns
    • If it is multilane, it should have the other lanes inhabiting the other r1_L* columns and r2_L* columns
  • Essentially, multilane files will be concatenated together into their respective reads and fed into the pipleine accordingly

The csv file will be passed into the pipeline using the --sample_table flag. Make sure to pass in the absolute path of the sample table.

Step 3: Build Singularity image file

• Run sh ./Bisulfite_Nextflow/scripts/singularity_defs/build_sif.sh
• This will allow for the Nextflow pipeline to run within the singularity image, controlling for dependencies.

Step 4: Choose genome that the samples will be aligned to

• This step requires you to have already downloaded your reference genome and prepared it using bismark_genome_preparation
• Once prepared, you can provide the folder of the genome to the --db flag. Make sure to pass in the absolute path of the genome folder.

Step 5: Run pipeline

Once you have your sample table and genome added to your directory, you can run the pipline using Nextflow.

Example run: nextflow run ./Bisulfite_Nexflow/main.nf -resume -with-report -with-trace -with-dag flowchart.mmd --db path_to_genome --sample_table path_to_sample_table

I usually always include the -resume flag when starting a pipeline run since it will just say nothing to be resumed if a cache is not found.

Additional flags

--rrbs: Passes flags to certain tools when the data is Reduced Represetation Bisulfite Sequencing (RRBS).
--multiqc_report_title: Title your multiqc report to be a bit more descriptive. Make sure not to include spaces in the title.

Dependencies:

Tasklist

• List dependencies
• Create Singularity/Docker Container file for the pipeline