Project name: gigas-WGBS-ploidy-desiccation
Funding source: unknown
Species: Crassostrea gigas
variable: ploidy, desiccation, high temperature

next notebook entry


We have bisulfide sequencing data from Ronit’s desiccation exposure experiments using juvenile pacific oysters. Here is the forked github repo.

List of the progress so far:

  1. Received WGBS data from ZymoResearch
  2. Files were added to the owl server
  3. Sam ran FastQC. Here is the multiQC report

Here is a list of samples

SeqID Library Name Tissue Ploidy Desiccation Heat Stress
zr3534_1 D11-C ctenidia diploid yes no
zr3534_2 D12-C ctenidia diploid yes no
zr3534_3 D13-C ctenidia diploid yes no
zr3534_4 D19-C ctenidia diploid yes yes
zr3534_5 D20-C ctenidia diploid yes yes
zr3534_6 T11-C ctenidia triploid yes no
zr3534_7 T12-C ctenidia triploid yes no
zr3534_8 T13-C ctenidia triploid yes no
zr3534_9 T19-C ctenidia triploid yes yes
zr3534_10 T20-C ctenidia triploid yes yes

Desiccation - desiccation for 24 hr at 27C,
Heat_stress - 1 hr at 45C

Pathway forward:

Now that we have the WGBS files and FastQC didn’t find any large errors, the next steps is to run Bismark. The files are pretty big, so instead of running it locally, I will use the resources of our hyak_mox server. Bismark performs alignments of bisulfite-treated reads to a reference genome and cytosine methylation calls at the same time.

The steps I will be following during this analysis are:

  1. logging into mox server
  2. Generating slurm script (.sh)

Step 1: Logging into mox


Step 2: slurm script and job scheduler

Slurm is an open source, fault-tolerant, and highly scalable cluster management and job scheduling system for large and small Linux clusters. Useful information can be found in the wiki and this example that Steven provided. Yaamini has also outlined a great pipeline for analysis and reporting.

To configure the job, I first had to develop my file structure:

mkdir -p mneorge/{analyses,blastdb,data,jobs,programs,sbatch_scripts}

parent folder: /gscratch/srlab/mngeorge
contents: analyses, blastdb, data, jobs, programs, sbatch_scripts

After completed, I copied ronit’s data to my data folder:

cp -avr /gscratch/srlab/sr320/data/cg /gscratch/srlab/mngeorge/data/cgigas_ronit

I then recreated a sbatch file within the sbatch_scripts ( subfolder to run bismark on the data.

Here is the code:

  GNU nano 2.3.1                          File:

  ## Job Name
  #SBATCH --job-name=ronit-bismark
  ## Allocation Definition
  #SBATCH --account=srlab
  #SBATCH --partition=srlab
  ## Nodes
  #SBATCH --nodes=1
  ## Walltime (days-hours:minutes:seconds format)
  #SBATCH --time=15-00:00:00
  ## Memory per node
  #SBATCH --mem=100G
  #SBATCH --mail-type=ALL
  ## Specify the working directory for this job
  #SBATCH --chdir=/gscratch/srlab/mngeorge/20210316-cgigas-ploidy-stress-bismark/

  # Directories and programs
  source /gscratch/srlab/mngeorge/sbatch_scripts/

  ${bismark_dir}/bismark_genome_preparation \
  --verbose \
  --parallel 28 \
  --path_to_aligner ${bowtie2_dir} \

  # /zr3644_11_R2.fastp-trim.20201206.fq.gz

  find ${reads_dir}*_R1.fastp-trim.20201202.fq.gz \
  | xargs basename -s _R1.fastp-trim.20201202.fq.gz | xargs -I{} ${bismark_dir}/bismark \
  --path_to_bowtie ${bowtie2_dir} \
  -genome ${genome_folder} \
  -p 8 \
  -score_min L,0,-0.6 \
  --non_directional \
  -1 ${reads_dir}{}_R1.fastp-trim.20201202.fq.gz \
  -2 ${reads_dir}{}_R2.fastp-trim.20201202.fq.gz \

  find *.bam | \
  xargs basename -s .bam | \
  xargs -I{} ${bismark_dir}/deduplicate_bismark \
  --bam \
  --paired \

  ${bismark_dir}/bismark_methylation_extractor \
  --bedGraph --counts --scaffolds \
  --multicore 28 \
  --buffer_size 75% \

  # Bismark processing report


  #Bismark summary report


  #run multiqc
  /gscratch/srlab/programs/anaconda3/bin/multiqc .

  # Sort files for methylkit and IGV

  find *deduplicated.bam | \
  xargs basename -s .bam | \
  xargs -I{} ${samtools} \
  sort --threads 28 {}.bam \

After saving, I then added to the queue:


You can check the position in the queue using squeue:

As well as the job status using the job number:

scontrol show job 1740292

Once the job is done, the resulting files should be stored on our lab server gannet. Here is an example script using rsync:

  ssh <login>
  cd /volume2/web/panopea/030521-ronrosM/
  rsync -avz --progress --verbose 030521-ronrosM