# Task: Extract GeneID-Number and gene information

#!/usr/bin/perl
use strict;
use warnings;

my $in;
my $data;
my @array;
my $array;
my $GeneID;
my @BMB;
my $BMB;
my $flag = 0;
my %hash;
my $hash;


# 1) open the .gff Inputfile and while reading line by line split $data at each tab and put them in the @array


open $in, '<', "Genomteil.gff" or die $!;

while ($data = <$in>) {

@array = split(/\t/, $data);

if ($array [2] =~/gene/){ #if you find the word 'gene' a textbloxk follows which contains some information I want to extract and put in an array)

$flag = 1;

@BMB = ($array[3], $array[4], $array[6]); #the array will be used as values for my hash later


}



if ($array[2] =~/CDS/){

push (@BMB, $array[2]);

#put more data in my array

} elsif ($array[2] =~/exon/){

push (@BMB, $array[2]);

}


if ($array[8] =~ /.*;db_xref=GeneID:(\d+)\n/) { #if you find the word 'GeneID' extract the following number and put it in my hash (as key), then put the array in my hash

$GeneID = $1;

@{$hash {$GeneID}} = @BMB;

}



if ($array [8]=~ /.*;exon_number=1/){ #if you find the word 'exon number', then the textblock is over
$flag = 0;

}

}

close $in;

while ( ($GeneID, $BMB) = each %hash) {
print "$GeneID => $BMB[0]\n";
}