collapse_columns.pbs

# processors on 1 node
#PBS -l nodes=1:ppn=3

#Request walltime/fastq file ~30 min/sample
#PBS -l walltime=240:00:00

#Request 1 gigabyte of memory per process
#PBS -l mem=16gb

#Request that regular output and terminal output go to the same file
#PBS -j oe
#PBS -M Emily.Laplante@bcm.edu
#PBS -m abe
#PBS -N Collapse_cellLines

#mkdir /scratch/$PBS_JOBID
cd /scratch/$PBS_JOBID

cp /mnt/brlstor/Vol6_SP/exrna/readCoverage/EmilyTest/biosample_RBP_out_unzipped/cellLines_all_samples_filtered.tar.gz /scratch/$PBS_JOBID

tar -xzvf cellLines_all_samples_filtered.tar.gz
rm *.tar.gz

sed 's/ /\t/g' cellLines_all_samples_filtered.txt > intermediate.txt
mv intermediate.txt cellLines_all_samples_filtered.txt

#remove first 3 rows
cut -d$'\t' -f 4- cellLines_all_samples_filtered.txt > noRowNames.txt
cut -d$'\t' -f 1-3 cellLines_all_samples_filtered.txt > rownames.txt
rm cellLines_all_samples_filtered.txt

echo -n "" > names.txt

for i in {0..76}; do #Should be 0 to 1 less than total number of samples
  let start=i*150+1
  let end=i*150+150
  echo "${start} ${end}"
  cut -d$'\t' -f ${start}-${end} noRowNames.txt >${i}_sep.txt
  
  sample_name=$(head -n 1 ${i}_sep.txt | sed 's/_RBP.*//g')
  echo ${sample_name}
  
  awk '{c=0;for(i=1;i<=NF;++i){c+=$i};print c}' ${i}_sep.txt > sample_total.txt
  if [ $sample_name == "EXR-AKRIC1AKGBM001-BS" ]; then
   #echo "equal"
   mv sample_total.txt study_total.txt
  else
   #echo "Not"
   paste study_total.txt sample_total.txt > study_total_temp.txt
   mv study_total_temp.txt study_total.txt
   rm sample_total.txt
  fi
  echo $sample_name >> names.txt
  rm ${i}_sep.txt
done

rm noRowNames.txt

NAMES_USE=$(sed 's/\n/ /g' names.txt)
echo $NAMES_USE | cat - study_total.txt > study_total_temp.txt && mv study_total_temp.txt study_total.txt
#concatenate columns 1-3
paste -d "\t" rownames.txt study_total.txt | sed 's/ /\t/g' > cellLines_all_summed_filtered.txt

##for some reason the first row is all zeros and there is one more row than rowname - so remove extra first line and correctly align regions
head -1 cellLines_all_summed_filtered.txt >names_use.txt
awk NR\>1 cellLines_all_summed_filtered.txt > test_rl_noname.txt
cut -d$'\t' -f 4- test_rl_noname.txt | sed '$d' | awk NR\>1 > test_rl_noRowNames.txt
cut -d$'\t' -f 1-3 test_rl_noname.txt | sed '$d' > test_rl_rownames.txt
tail -1 cellLines_all_summed_filtered.txt | tail -c +2 > test_lastLine.txt

cat test_rl_noRowNames.txt test_lastLine.txt > total_read.txt
paste test_rl_rownames.txt total_read.txt | sed 's/ /\t/g' > total_fixed_values.txt
cat names_use.txt total_fixed_values.txt > total_mat.txt
mv total_mat.txt cellLines_all_summed_filtered.txt


rm study_total.txt

tar -czvf cellLines_all_summed_filtered.tar.gz cellLines_all_summed_filtered.txt

mv cellLines_all_summed_filtered.tar.gz /mnt/brlstor/Vol6_SP/exrna/readCoverage/EmilyTest/biosample_RBP_out_unzipped/