十年的生信

长数据如下图

perl代码如下

#!/usr/bin/perl -w
use strict;
use warnings;
use Data::Dumper;

my $usage=<<USAGE;
Usage:
    perl $0 inputfile
USAGE
if(@ARGV==0){die $usage};

my $file=$ARGV[0];
my %timeinfo=();
my $reftime=\ %timeinfo;
my $i=0;

open(RF,$file) || die $!;
open(WF,">kuan_data.txt") || die $!;
while(my $line=<RF>){
    next if ($.==1);
    chomp($line);
    my @arr=split('\t',$line);
    $reftime -> {$arr[0]} -> {$arr[1]} =$arr[2];

}

for my $j (keys %{$reftime}){
    $i=0;
    for my $dd (keys %{$reftime -> {$j}}){
        $i++;
    }
}

for my $year (sort {$a cmp $b} keys %{$reftime}){ 
    my $j=1;
    for my $month (sort {$a <=> $b} keys %{$reftime -> {$year}}){  
        if ($j==1){
            print WF $month,"\t";
            $j++;
        }elsif($j<$i){
            print WF $month,"\t";
            $j++;
        }elsif($j==$i){
            print WF $month,"\n";
        }else{
            next;
        }
    }
    last;
}


for my $year (sort {$a cmp $b} keys %{$reftime}){
    my $j=1;
    for my $month (sort {$a <=> $b} keys %{$reftime -> {$year}}){
        if ($j==1){
            print WF $year,"\t",$reftime -> {$year} -> {$month},"\t";
            $j++;
        }elsif($j>0 && $j < $i){
            print WF $reftime -> {$year} -> {$month},"\t";
            $j++;
        }else{
            print WF $reftime -> {$year} -> {$month},"\n";
        }
    }
}

close(RF);
close(WF);

得到最后的宽数据，如下图

首先输入的文件类似如下图 为了转换为长数据，编写如下perl代码

#!/usr/bin/perl -w
use strict;
use warnings;
use Data::Dumper;

my $file=$ARGV[0];
my @matrix=();
my @month=();
my @year=();

open(RF,$file) || die $!;
open(WF,">tidy_data.txt") || die $!;

my $i=0;
while (my $line=<RF>){
    next if ($.==1);
    chomp($line);
    my @arr=split('\t',$line);
    shift(@arr);
    for my $j (0..@arr-1){
        $matrix[$i][$j]=$arr[$j];
    }
    $i++;
}
close(RF);

open(RF,$file) || die $!;
open(WFF,">process.txt") || die $!;

while(my $line=<RF>){
    if ($.==1){
        chomp($line);
        @month=split('\t',$line);
        next;
    }
    chomp($line);
    my @arr=split('\t',$line);
    print WFF $arr[0],"\n";
}
close(RF);
close(WFF);

open(RFF,"process.txt") || die $!;
while(my $line=<RFF>){
    chomp($line);
    push @year,$line;
}
close(RFF);

for my $i (0..$#year){
    for my $j (0..$#month){
        print WF $year[$i],"\t",$month[$j],"\t",$matrix[$i][$j],"\n";
    }
}

close(WF);
system("del process.txt")

便可实现转换，结果如下图