flowr tips

bioawk on flowr status

# install this cool tools from Heng Li (https://github.com/lh3/bioawk)
brew install bioawk

# information regarding exit_code:
# - NA: not started yet

# show lines where job started and had errors
bioawk -tc hdr '{if($exit_code!=0 && $started=="TRUE") print $jobname, $cmd, $exit_code}' flow_details.txt

# show lines which did not start
bioawk -tc hdr '{if($started=="FALSE") print $jobname, $cmd, $exit_code}' flow_details.txt

# started, but have not finished yet (no exit status yet)
bioawk -tc hdr '{if($started=="TRUE" && $exit_code==-1) print $jobname, $cmd, $exit_code}' flow_details.txt

# get the 3rd column, cut -f3 and re-submit the jobs
jobs=$(bioawk -tc hdr '{if($exit_code!=0 && $started=="TRUE") print $jobname, $cmd, $exit_code}' flow_details.txt | cut -f2)
echo $jobs
# this will work, provided there were no pre requisite jobs
for j in $jobs; do bsub < $j; done











# END
 @sahilseth
Owner
sahilseth commented on Aug 7, 2017
 # resubmission ex
 bsub -M 60384 -R rusage[mem=60384] < cnt_cmd_66.sh

bioawk_on_genomic_fls.sh

# https://github.com/stephenturner/oneliners
# http://www.delorie.com/gnu/docs/gawk/gawk_204.html
vcffl="sample.vcf"


bioawk -c vcf '{
    freq[$filter]++
    total++
  } 
  END {
    for(val in freq) 
      printf "%s\t%d\t%d\n", val, freq[val], freq[val]*100/total
      }' $vcffl | sort -k 2nr | less

change file content

# https://stackoverflow.com/questions/10445934/change-multiple-files

fls=$(ls */002.cmd_hla/cmd_hla_cmd_1.sh)


for fl in $fls; do
    # test first
    #sed 's/Allelelist.txt/Allelelist_v2.txt/g' $fl
    sed -i 's/Allelelist.txt/Allelelist_v2.txt/g' $fl
done

flowr tips

readme

flowr_run_with_uuid.sh

# install the newer version:
flowr devtools::install_github repo=sahilseth/flowr ref=add_docker

# create a UUID path
myuuid=$(flowr get_unique_id prefix=tumorname___normalname)

# run an example, with predefined uuid
flowr run x=sleep_pipe uuid=$myuuid

gistfile1.txt

# get a list of rundirs
rundirs=$(ls . | grep neo);echo $rundirs

jobnm=""
jobfl=""

for rundir in $rundirs; do
  echo "working on $rundir"
  # cd into tmp, to run the cmd
  cd $rundir/tmp
  # extract and run the cmd
  grep "Allelelist_v2" ../$jobnm/$jobfl | bash
  cd ../..
done