These are a pysano
cmd.txt
templates for running jobs on the HCI nodes at CHPC. These
could be converted to a slurm script for direct execution on CHPC as necessary by
replacing paths and adding headers.
Carefully check the header lines at the top of the templates, and make adjustments as necessary (email, etc).
Contact the Core if you have any issues.
There are two workflows here: Novoalign/USeq and Bismark.
Novoalign currently gives a better alignment rate than Bismark/Bowtie2, at the expense of speed. With moderate-sized sequencing projects, you may still time out (10 days) on a single node.
Therefore, split your Fastq reads into a few million reads apiece, and align each job independently on separate nodes (You will thank me later). Split the file manually, or run this script, which maintains compression and executes with multiple threads.
Run the novoalign_part_cmd.txt
alignment on each part. When finished, merge the
resulting bam files into a single bam file with samtools and index.
Once you have a single file, you can run the useq_bisulfite_cmd.txt
job with the
sample bam file.
You will need to run additional USeq applications for calling differentially methylated regions. See the USeq documentation.
NOTE The cmd.txt
template files are organism specific. Add or remove the comment marks
as necessary to match your organism genome version. Contact the Core if the genome
version you need doesn't exist.
An example Bismark alignment and initial processing steps are provided in the
bismark_cmd.txt
template file.
This will use the GNomEx sample identifier (e.g. 1234X1) from the Fastq file name as the base name for the output files.
Note the organism parameter at the top and adjust accordingly.