Streamline getting expression data for a particular array of genes in a particular brain tissues

gen.CSV.perTissue.sh

#This is a shellscript intended to streamline the process of checking an initial array of genes against
#1) a file storing relevant column numbers for a particular neural tissue,
#2) a file storing expression data regarding that tissue and those genes

#!/usr/bin/perl

use strict;
#use warnings;

require "getopts.pl";
   
use vars qw ($opt_f $opt_t $opt_s $opt_h $opt_H);

&Getopts('f:t:s:hH:');
my $executable = "gen.CSV.perTissue.pl";
$| = 1;
my $usage = qq{
    
  ###################################################################
  usage:  $executable -f fileName
          used for transposing a table (rows to columns and columns to rows)

	  # -f the file name of the array of genes of interest; will use stdin if specified as "stdin".
          #  
          # -t the tissue of interest, i.e. "hippocampus" or "dorsal prefrontal cortex"
          # 
          # -s column separator
          #    [default: tab]
          #    C: comma
          #    S: space   
          #
	  # -hH print this 
	example: >$executable -f testInputFile -s S > output

 ####################################################################
	\n};
my ($genesOfInterest, $tissueOfInterest, $expressionData);
# options:
if (length($opt_h) > 0 || length($opt_H) > 0 ) {die($usage);}
if (-s $opt_f) {$genesOfInterest = $opt_f; } 
elsif ($opt_f =~ /stdin/i) {$genesOfInterest = "stdin"; }
else {die("Error: must have a input file for -f $opt_f !!!\n$usage"); }
if (-s $opt_t) {$tissueOfInterest = $opt_t}
if ($tissueOfInterest =~ /" "/) {$tissueofInterest =~ s/" "/_/}
#if ($opt_s =~ /^C/i ) {$separator = "comma";  }elsif ($opt_s =~ /^S/i) {$separator = "space"; } else {$separator = "tab"; }






######This command merges two tables together on the basis of common genes found in one table, then stores these in a new file


rosseraa@localhost:~/work.dir>/mnt/xgen.programs/qyScripts/merge.basedOn.column.pl -f BrainSPAN/RNAseq.genes.RPKM/rows_metadata.csv.gz -l 4 -F 0 -L '"' -q arrayTable.txt -m 1 -Q 0 -s C -t T > "$genesOfInterest.arrayTable.rowNum"






######These commands added 1 to all the values returned for columns that contained the string "hippocampus" 


rosseraa@localhost:~/work.dir>z BrainSPAN/RNAseq.genes.RPKM/columns_metadata.csv.gz 
| grep -i "$tissueOfInterest" |cut -d"," -f1 
| ../Documents/practice/transposer.pl -f stdin  
| perl -ne '{chomp; @a =split (/\t/, $_); foreach $a (@a) {$b = $a + 1;  print "$b,"; }  print "\n" if (eof); }'  


rosseraa@localhost:~/work.dir>z BrainSPAN/RNAseq.genes.RPKM/expression_matrix.csv.gz 
|cut -d"," -f"$genesOfInterest.arrayTable.rowNum" > "$expressionData.data.txt"




########Last step: This command uses the merge script to concatenate table data and store that output in a new file


rosseraa@localhost:~/work.dir>/mnt/xgen.programs/qyScripts/merge.basedOn.column.pl -f "$expressionData" -l 1 -F 0 -q "$genesOfInterest.arrayTable.rowNum" -m 1 -Q 0 -s C -t C  |cut -f1  > "$tissueOfInterest.data.4arrayGenes.csv"