GSEA分析--结果解读

输入文件symbol.png

输入文件risk.png

use strict;

my %hash=();
my $normalFlag=0;

open(RF,"risk.txt") or die $!;
while(my $line=<RF>){
    chomp($line);
    my @arr=split(/\t/,$line);
    $hash{$arr[0]}=$arr[$#arr-1];
}
close(RF);

my @indexs=();
my @riskArr=();
open(RF,"symbol.txt") or die $!;
open(WF,">sampleSymbol.txt") or die $!;
my $normalCount=0;
my $tumorCount=0;
while(my $line=<RF>){
    chomp($line);
    my @arr=split(/\t/,$line);
    if($.==1){
        print WF "ID";
        for(my $i=1;$i<=$#arr;$i++){
            my @samples=split(/\-/,$arr[$i]);
            if($samples[3]=~/^0/){
                my $sampleName="$samples[0]-$samples[1]-$samples[2]";
                if(exists $hash{$sampleName}){
                        push(@indexs,$i);
                        push(@riskArr,$hash{$sampleName});
                        print WF "\t$arr[$i]";
                        $tumorCount++;
                        delete($hash{$sampleName});
                }
            }
        }
        print WF "\n";
    }
    else{
        print WF $arr[0];
        foreach my $col(@indexs){
            print WF "\t$arr[$col]";
        }
        print WF "\n";
    }
}
print WF "Risk\t" . join("\t",@riskArr) . "\n";
close(WF);
close(RF);


my $colNum=0;
my $rowNum=0;
%hash=();
my $geneName="Risk";
@indexs=();
my @geneArr=();

open(RF,"sampleSymbol.txt") or die $!;
while(my $line=<RF>){
    chomp($line);
    my @arr=split(/\t/,$line);
    if($.==1){
        for(my $i=1;$i<=$#arr;$i++){
            my @samples=split(/\-/,$arr[$i]);
            if($samples[3]=~/^0/){
              push(@indexs,$i);
              my $sampleName=$arr[$i];
              $hash{$sampleName}=1;
              $colNum++;
            }
        }
    }
    else{
            $rowNum++;
          if($arr[0] eq $geneName){
              foreach my $col(@indexs){
                  push(@geneArr,$arr[$col]);
              }
          }
    }
}
close(RF);

my $firstGeneVal=$geneArr[0];
my $geneMed=median(@geneArr);

open(RF,"sampleSymbol.txt") or die $!;
open(WF,">$geneName.gct") or die $!;
print WF "#1.2\n";
print WF "$rowNum\t$colNum\n";
open(CLS,">$geneName.cls") or die $!;
print CLS "$colNum\t2\t1\n";
my @samp1e=(localtime(time));
if($firstGeneVal>$geneMed){
    print CLS "#\th\tl\n";
}
else{
    print CLS "#\tl\th\n";
}

@indexs=();
my @typeArr=();
while(my $line=<RF>){
    chomp($line);
    my @arr=split(/\t/,$line);
    if($.==1){
        print WF "NAME\tDESCRIPTION";
        for(my $i=1;$i<=$#arr;$i++){
            my @samples=split(/\-/,$arr[$i]);
            if($samples[3]=~/^0/){
              my $sampleName=$arr[$i];
              if(exists $hash{$sampleName}){
                  push(@indexs,$i);
                  print WF "\t$arr[$i]";
                  #delete($hash{$sampleName});
              }
            }
        }
        print WF "\n";
    }
    else{
        my $symbolName=$arr[0];
        $symbolName=~s/(.+?)\|.+/$1/g;
        print WF "$symbolName\tna";
        foreach my $col(@indexs){
            print WF "\t$arr[$col]";
        }
            print WF "\n";
        if($arr[0] eq $geneName){
            foreach my $col(@indexs){
                if($arr[$col]>$geneMed){
                  push(@typeArr,"h");
              }
              else{
                push(@typeArr,"l");
              }
            }
        }
    }
}
print CLS join("\t",@typeArr) . "\n";
close(WF);
close(CLS);
close(RF);
unlink<"sampleSymbol.txt">;

sub median{  
    my (@data) = sort {$a <=> $b} @_;  #将原数组重新排序  
    if(scalar (@data) % 2){  
        return ($data [@data / 2]);  #@data/2的返回值向上取整  
    }else {  
        my ($upper ,$lower);  
        $upper = $data[@data / 2];  
        $lower = $data[@data / 2 -1];  
        return (($lower+$upper) / 2);  
    }  
}  

输出文件为：

Risk.gct.png

Risk.cls.png

下一步输入文件准备：

输入文件准备.png

下载GSEA软件.png

加载数据

软件界面.png

选择分析模式.png

决定GSEA富集图在左边还是右边.png

是否需要gene symbol注释.png

最终界面，点击run运行.png

软件分析完后找到图中文件.png