RNAseq cufflinks로 나온 genes.fpkm_tracking파일 합치기

#!/bin/bash

# This script is to quantify samsung's RNAseq data

#$ -S /bin/bash

#$ -N bulkCombine

#$ -cwd

#========================================

# Set up parameters

#========================================

out_dir=/path_to_FPKM_file_dir/

suffix_1=/genes.fpkm_tracking

com_dir=/path_to_FPKM_dir_header/header_dir

#=======================================

# Get file prefix

#=======================================

ls | sort | uniq | grep -v .sh | grep -v .txt | grep -v bulkCombine | grep -v tmp > fpkm_prefix.txt # Get list of directories having genes.fpkm_tracking

cut -f1-7 $com_dir$suffix_1 | sort -rn -k1,1 > GBM_bulkRseq_validation_20180521.txt

# Make and sort the header file according to ENSG id

#======================================

# Do combinining the FPKM values by for-looping

#======================================

for input in $(cat fpkm_prefix.txt);

do

        echo $input

        cut -f1 GBM_bulkRseq_validation_20180521.txt > tmp1

        cut -f1 $out_dir$input$suffix_1 | sort -rn -k 1,1 | cut -f1 > tmp2

        diff tmp1 tmp2 # Check whether sorting was properly conducted or not.

        cat $out_dir$input$suffix_1 | sort -rn -k 1,1 | cut -f10 | sed "s/FPKM/$input/" > tmp # Replace FPKM into sample name

        paste GBM_bulkRseq_validation_20180521.txt tmp > tmp2 # Column binding

        mv tmp2 GBM_bulkRseq_validation_20180521.txt

done