Perl program to look for k-mer with specific sequence - perl

Question

I am trying to enhance a perl program I have previously written so that it recognizes top 1000 length 23 k-mers that ends with GG and print out the k-mers that only appears once in the sequence. However, no matter where I add the reg exp, I am unable to get the expected result.
The code I have:
#!/usr/bin/perl
use strict;
use warnings;
my $k = 23;
my $input = 'Fasta.fasta';
my $output = 'Fasta2.fasta';
my $match_count = 0;
#Open File
unless ( open( FASTA, "<", $input ) ) {
die "Unable to open fasta file", $!;
}
#Unwraps the FASTA format file
$/ = ">";
#Separate header and sequence
#Remove spaces
unless ( open( OUTPUT, ">", $output ) ) {
die "Unable to open file", $!;
}
<FASTA>; # discard 'first' 'empty' record
my %seen;
while ( my $line = <FASTA> ) {
chomp $line;
my ( $header, #seq ) = split( /\n/, $line );
my $sequence = join '', #seq;
for ( length($sequence) >= $k ) {
$sequence =~ m/([ACTG]{21}[G]{2})/g;
for my $i ( 0 .. length($sequence) - $k ) {
my $kmer = substr( $sequence, $i, $k );
##while ($kmer =~ m/([ACTG]{21}[G]{2})/g){
$match_count = $match_count + 1;
print OUTPUT ">crispr_$match_count", "\n", "$kmer", "\n" unless $seen{$kmer}++;
}
}
}
The input fasta file looks like this:
> >2L type=chromosome_arm; loc=2L:1..23011544; ID=2L; dbxref=REFSEQ:NT_033779,GB:AE014134; MD5=bfdfb99d39fa5174dae1e2ecd8a231cd; length=23011544; release=r5.54; species=Dmel;
CGACAATGCACGACAGAGGAAGCAGAACAGATATTTAGATTGCCTCTCAT
TTTCTCTCCCATATTATAGGGAGAAATATGATCGCGTATGCGAGAGTAGT
GCCAACATATTGTGCTCTTTGATTTTTTGGCAACCCAAAATGGTGGCGGA
TGAACGAGATGATAATATATTCAAGTTGCCGCTAATCAGAAATAAATTCA
TTGCAACGTTAAATACAGCACAATATATGATCGCGTATGCGAGAGTAGTG
CCAACATATTGTGCTAATGAGTGCCTCTCGTTCTCTGTCTTATATTACCG
CAAACCCAAAAAGACAATACACGACAGAGAGAGAGAGCAGCGGAGATATT
TAGATTGCCTATTAAATATGATCGCGTATGCGAGAGTAGTGCCAACATAT
TGTGCTCTCTATATAATGACTGCCTCTCATTCTGTCTTATTTTACCGCAA
ACCCAAATCGACAATGCACGACAGAGGAAGCAGAACAGATATTTAGATTG
CCTCTCATTTTCTCTCCCATATTATAGGGAGAAATATGATCGCGTATGCG
AGAGTAGTGCCAACATATTGTGCTCTTTGATTTTTTGGCAACCCAAAATG
GTGGCGGATGAACGAGATGATAATATATTCAAGTTGCCGCTAATCAGAAA
TAAATTCATTGCAACGTTAAATACAGCACAATATATGATCGCGTATGCGA
GAGTAGTGCCAACATATTGTGCTAATGAGTGCCTCTCGTTCTCTGTCTTA
TATTACCGCAAACCCAAAAAGACAATACACGACAGAGAGAGAGAGCAGCG
GAGATATTTAGATTGCCTATTAAATATGATCGCGTATGCGAGAGTAGTGC
CAACATATTGTGCTCTCTATATAATGACTGCCTCTCATTCTGTCTTATTT
TACCGCAAACCCAAATCGACAATGCACGACAGAGGAAGCAGAACAGATAT
and so on...
The expected outcome (print out the 23k-mers with GG ending that only appear once in the sequence) I am hoping to get:
>crispr_1
GGGTGGAGCTCCCGAAATGCAGG
>crispr_2
TTAATAAATATTGACACAGCGGG
>crispr_3
ATCGTGGGGCGTTTTGTGAAAGG
>crispr_4
AGTTTTTCACATAATCAGACAGG
>crispr_5
GTGTTGGATGAGTGTCCTCTGGG
>crispr_6
ATAGGTTGGTTGTTTTAAAAGGG
>crispr_7
AAATTTTTGTTGCCACTGAATGG
>crispr_8
AAGTTTCGAACTACGATGGTTGG
>crispr_9
CATGCTTTGTGGAAATAAGTCGG
>crispr_10
CACAGTGGGTGTTTGCACCTCGG
.... and so on
The current code I did create a fasta file with following:
>crispr_1
CGACAATGCACGACAGAGGAAGC
>crispr_2
GACAATGCACGACAGAGGAAGCA
>crispr_3
ACAATGCACGACAGAGGAAGCAG
>crispr_4
CAATGCACGACAGAGGAAGCAGA
>crispr_5
AATGCACGACAGAGGAAGCAGAA
>crispr_6
ATGCACGACAGAGGAAGCAGAAC
>crispr_7
TGCACGACAGAGGAAGCAGAACA
>crispr_8
GCACGACAGAGGAAGCAGAACAG
>crispr_9
CACGACAGAGGAAGCAGAACAGA
>crispr_10
ACGACAGAGGAAGCAGAACAGAT
.... and so on
while if I remove the
for (length($sequence) >=$k){
$sequence =~m/([ACTG]{21}[G]{2})/g;
and add the ##while ($kmer =~ m/([ACTG]{21}[G]{2})/g){
while ($kmer =~ m/([ACTG]{21}[G]{2})/g){
I am getting fasta file (with results which is not numbered correctly and unable to distinguish between duplicated and unique sequences):
>crispr_1
CATTTTCTCTCCCATATTATAGG
>crispr_2
ATTTTCTCTCCCATATTATAGGG
>crispr_3
TATTGTGCTCTTTGATTTTTTGG
>crispr_4
GATTTTTTGGCAACCCAAAATGG
>crispr_5
TTTTTGGCAACCCAAAATGGTGG
>crispr_6
TTGGCAACCCAAAATGGTGGCGG
>crispr_7
ACGACAGAGAGAGAGAGCAGCGG
>crispr_8
AAATCGACAATGCACGACAGAGG
>crispr_91
TATTGTGATCTTCGATTTTTTGG
>crispr_93
TTTTTGGCAACCCAAAATGGAGG
.... and so on
I have attempted to move the regex around my code, but none of them generated the expected result. I do not know what I did wrong over here. I have not add the exit the program when count reaches 1000 into the code yet.
Thanks in advance!

Answer 1

I'm not sure I understand your question completely, but could the following be what you need.
<FASTA>; # discard 'first' 'empty' record
my %data;
while (my $line = <FASTA>){
chomp $line;
my($header, #seq) = split(/\n/, $line);
my $sequence = join '', #seq;
for my $i (0 .. length($sequence) - $k) {
my $kmer = substr($sequence, $i, $k);
$data{$kmer}++ if $kmer =~ /GG$/;
}
}
my $i = 0;
for my $kmer (sort {$data{$b} <=> $data{$a}} keys %data) {
printf "crispr_%d\n%s appears %d times\n", ++$i, $kmer, $data{$kmer};
last if $i == 1000;
}
Some output on a file I have is:
crispr_1
ggttttccggcacccgggcctgg appears 4 times
crispr_2
ccgagctgggcgagaagtagggg appears 4 times
crispr_3
gccgagctgggcgagaagtaggg appears 4 times
crispr_4
gcacccgggcctgggtggcaggg appears 4 times
crispr_5
agcagcgggatcgggttttccgg appears 4 times
crispr_6
gctgggcgagaagtaggggaggg appears 4 times
crispr_7
cccttctgcttcagtgtgaaagg appears 4 times
crispr_8
gtggcagggaagaatgtgccggg appears 4 times
crispr_9
gatcgggttttccggcacccggg appears 4 times
crispr_10
tgagggaaagtgctgctgctggg appears 4 times
crispr_11
agctgggcgagaagtaggggagg appears 4 times
. . . .
ggcacccgggcctgggtggcagg appears 4 times
crispr_50
gaatctctttactgcctggctgg appears 4 times
crispr_51
accacaacattgacagttggtgg appears 2 times
crispr_52
caacattgacagttggtggaggg appears 2 times
crispr_53
catgctcatcgtatctgtgttgg appears 2 times
crispr_54
gattaatgaagtggttattttgg appears 2 times
crispr_55
gaaaccacaacattgacagttgg appears 2 times
crispr_56
aacattgacagttggtggagggg appears 2 times
crispr_57
gacttgatcgattaatgaagtgg appears 2 times
crispr_58
acaacattgacagttggtggagg appears 2 times
crispr_59
gaaccatatattgttatcactgg appears 2 times
crispr_60
ccacagcgcccacttcaaggtgg appears 1 times
crispr_61
ctgctcctgggtgtgagcagagg appears 1 times
crispr_62
ccatatattatctgtggtttcgg appears 1 times
. . . .
Update
To get the results you mentioned in your comment (below), replace the output code with:
my $i = 1;
while (my ($kmer, $count) = each %data) {
next unless $count == 1;
print "crispr_$i\n$kmer\n";
last if $i++ == 1000;
}
To answer my own comment to get first 1000.
<FASTA>; # discard 'first' 'empty' record
my %seen;
my #kmers;
while (my $line = <FASTA>){
chomp $line;
my($header, #seq) = split(/\n/, $line);
my $sequence = join '', #seq;
for my $i (0 .. length($sequence) - $k) {
my $kmer = substr($sequence, $i, $k);
if ($kmer =~ /GG$/) {
push #kmers, $kmer unless $seen{$kmer}++;
}
}
}
my $i = 1;
for my $kmer (#kmers) {
next unless $seen{$kmer} == 1;
print "crispr_$i\n$kmer\n";
last if $i++ == 1000;
}
Answer To check for uniqueness of final 12 chars ending in GG, the code below achieves that.
if ($kmer =~ /(.{10}GG)$/) {
my $substr = $1;
push #kmers, $kmer unless $seen{$substr}++;
}
my $i = 1;
for my $kmer (#kmers) {
my $substr = substr $kmer, -12;
next unless $seen{$substr} == 1;
print "crispr_$i\n$kmer\n";
last if $i++ == 1000;
}

Answer 2

Actually, this code line
$sequence =~m/([ACTG]{21}[G]{2})/g;
this line is just for the regex match, if you try to print this $sequence, it will surely print out the original result.
please add the code segement like this
if($sequence =~/([ACTG]{21}[G]{2}$)/g)
{
}#please remember to match the end with $.
BTW,It looks like the multiple for loop to parse this data is not very reasonable, the parse speed is without the best-efficiency.