Dinh/Dinh 2013/NOTES/2013-1-7
Jump to navigation
Jump to search
bisReadMapper pipeline
Triton
- First map reads using a shell script:
- Edit Go.mapBisulfite.sh as follows:
### EDIT below #### # current directory: change the following path to your working directory. cur_dir="/oasis/triton/scratch/ddiep/Working/WGBS_Noi_hg19/HELLO" # reads directory: change the following path to the reads directory # reads_dir and cur_dir doesn't have to be the same reads_dir="/oasis/triton/scratch/ddiep/Working/WGBS_Noi_hg19/HELLO" # List the index names (or unique names to each read) INDX="Indx1 Indx2 Indx3" email="ddiep@ucsd.edu" mm=2 # max number of allowable mismatches qual=64 # base quality offset value qtrim=20 # Phred quality value for soft-trimming threetrim=0 # number of bases to trim from 3' end (rightmost) fivetrim=0 # number of bases to trim from 5' end (leftmost) p=8 ### DO NOT EDIT, unless you know where these files are #### # the following paths should stay the same on triton. bisReadMapper="/home/ddiep/scripts/MethylationPipeline/smartBisReadMapper.pl" template_fwd="/projects/zhang-lab/ddiep/LatestGenome/BisRef/bisHg19/hg19.fa.bis.fwd.index" template_rev="/projects/zhang-lab/ddiep/LatestGenome/BisRef/bisHg19/hg19.fa.bis.rev.index" template_fa="/projects/zhang-lab/ddiep/LatestGenome/BisRef/bisHg19/hg19.fa" soap="/home/ddiep/softwares/soap2.21release/soap" ############################################################ cd $cur_dir for n in ${INDX} do ### f="s_1_1_$n.txt" # name format for read 1, make sure it matches the file names g="s_1_2_$n.txt" # name format for read 2, make sure it matches the file names ### echo "#!/bin/csh" > $n.job echo "#PBS -q small" >> $n.job echo "#PBS -l nodes=1:ppn=8" >> $n.job echo "#PBS -l walltime=36:00:00" >> $n.job echo "#PBS -o $n.log" >> $n.job echo "#PBS -e $n.err" >> $n.job echo "#PBS -V" >> Idx$n.job echo "#PBS -M diep.hue.dinh@gmail.com" >> $n.job echo "#PBS -m abe" >> $n.job echo "#PBS -A zhang-lab" >> $n.job echo "cd $cur_dir" >> $n.job echo "$bisReadMapper -r $reads_dir/$f,$reads_dir/$g -m $mm -W $template_fwd -C $template_rev -g $template_fa -a $soap -b $qual -p $p -n $n -q $qtrim -5 $fivetrim -3 $threetrim > $n.status" >> $n.job echo "rm *encoded" >> $n.job qsub $n.job done
- Run the script:
sh Go.mapBisulfite.sh > job_ids
- Now, there will be a different job for each sequence file that you have.
- If you have multiple sequencing lanes, just change the f and g variables to s_2, s_3, etc, and re-run the shell script with the command above.