Thank you, sir.

$count = $line =~ tr/ACGT//;  # count only the ACGT characters in $lin
+e
$line =~ tr/ACGT//cd;         # remove all non-ACGT characters from $l
+ine
[download]

Ok guys, listen up. The implicit specification of the data given in the OP is that the base sequence to be counted is given in a single line of arbitrary length. The next specification bump will be to an arbitrary number of lines of arbitrary length for the base sequence. Be prepared.

Hello Peter Keystrokes,

I notices a few things on your code.

You have open (my $fh, '>', $filename) or die "Could not open '$filename' $!";. From the open documentation:

Simple examples to open a file for reading:

        open(my $fh, "<", "input.txt")
        or die "Can't open < input.txt: $!";

and for writing:

        open(my $fh, ">", "output.txt")
        or die "Can't open > output.txt: $!";
[download]

Also on your while loop you have while (my $line = <>){ without the file handler. From How can I read in an entire file all at once?:

while (<$input>) {
    chomp;
    # do something with $_
}
[download]

At the end of your code you are trying to write to a file print $fh $count;. You have opened the file in writing mode from the beginning but I am wondering how you are reading the file.

So I assume you meant to write something like that?

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

my $filename = 'counts.txt';

open (my $fh, '+<', $filename)
    or die "Could not open '$filename' $!";

my $count = "";

while (<$fh>) {
    chomp;
    if ($_ =~ /^>hsa/) {
    $count .= length $_;
        $count .= " ";
    }
}

print $fh $count . "\n";
close $fh
    or die "Could not close '$filename' $!";

__END__

>hsa_circ_0000001|chr1:1080738-1080845-|None|None
ATGGGGTTGGGTCAGCCGTGCGGTCAGGTCAGGTCGGCCATGAGGTCAGGTGGGGTCGGCCATGAAGGTG
+GTGGGGGTCATGAGGTCACAAGGGGGTCGGCCATGTG
>hsa_circ_0000002|chr1:1158623-1159348-|NM_016176|SDF4
GGTGGATGTGAACACTGACCGGAAGATCAGTGCCAAGGAGATGCAGCGCTGGATCATGGAGAAGACGGCC
+GAGCACTTCCAGGAGGCCATGGAGGAGAGCAAGACACACTTCCGCGCCGTGGACCCTGACGGGGACGGT
+CACGTGTCTTGGGACGAGTATAAGGTGAAGTTTTTGGCGAGTAAAGGCCATAGCGAGAAGGAGGTTGCC
+GACGCCATCAGGCTCAACGAGGAACTCAAAGTGGATGAGGAAA
49 54
[download]

Notice the open (my $fh, '+<', $filename) or die "Could not open '$filename' $!";. On this mode you can read and write to the file.

I do not know exactly why you want to read and write to your file, for me it would made more sense to read and print the output based on characters that they where counted but if this is what you desire.

I would prefer to do it like this, just personal preference nothing special.

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

my $filename = 'counts.txt';

open (my $fh, '+<', $filename)
    or die "Could not open '$filename' $!";

chomp(my @lines = <$fh>);

for (@lines) { print $fh length($_) . " " if ($_ =~ /^>hsa/); }

close $fh
    or die "Could not close '$filename' $!";

__END__

>hsa_circ_0000001|chr1:1080738-1080845-|None|None
ATGGGGTTGGGTCAGCCGTGCGGTCAGGTCAGGTCGGCCATGAGGTCAGGTGGGGTCGGCCATGAAGGTG
+GTGGGGGTCATGAGGTCACAAGGGGGTCGGCCATGTG
>hsa_circ_0000002|chr1:1158623-1159348-|NM_016176|SDF4
GGTGGATGTGAACACTGACCGGAAGATCAGTGCCAAGGAGATGCAGCGCTGGATCATGGAGAAGACGGCC
+GAGCACTTCCAGGAGGCCATGGAGGAGAGCAAGACACACTTCCGCGCCGTGGACCCTGACGGGGACGGT
+CACGTGTCTTGGGACGAGTATAAGGTGAAGTTTTTGGCGAGTAAAGGCCATAGCGAGAAGGAGGTTGCC
+GACGCCATCAGGCTCAACGAGGAACTCAAAGTGGATGAGGAAA
49 54
[download]

Update 3: Thanks to Anonymous Monk, AnomalousMonk, and Laurent_R I finally understand my mistakes and this is the updated recommended version.

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

my $count;
while (<>) {
    if ($_ =~ /^>hsa/){
    chomp (my $line = <>);
    $count .= length($line);
        $count .= " ";
    print "Count final: " . $count . "\n";
    }
} continue {
    close ARGV if eof; # reset $. each file
}

my $filename = 'counts.txt';

open (my $fh, '>', $filename)
    or die "Could not open '$filename' $!";

print $fh $count;

close $fh
    or die "Could not close '$filename' $!";

__END__

$ perl test.pl test.txt
Count final: 107
Count final: 107 251
[download]

And the way that I would try to resolve it.

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

my $filename = 'counts.txt';

open (my $fh, '>', $filename)
    or die "Could not open '$filename' $!";

chomp(my @lines = <>);

for (@lines) {
    next if ($_ =~ /^>hsa/);
    print $fh length($_) . " ";
}

print $fh "\n";

close $fh
    or die "Could not close '$filename' $!";

__END__

$ cat counts.txt
107 251
[download]

Update: 4 Even better as Anonymous Monk point it out:

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

my $filename = 'counts.txt';

open (my $fh, '>', $filename)
    or die "Could not open '$filename' $!";

chomp(my @lines = <>);

for (@lines) { print $fh length($_) . " " if ($_ =~ tr/ACGT//); }

print $fh "\n";

close $fh
    or die "Could not close '$filename' $!";

__END__

$ cat counts.txt
107 251
[download]

Hope this helps.

Hi thanos1983. Some comments on your post which I hope may avoid future confusion.

You have open (my $fh, '>', $filename) or die "Could not open '$filename' $!";. From the open documentation ...

Except for the use of  $filename in the OPed code instead of repeated literal strings (which I think is good practice in the interest of DRYness) and the inclusion of the mode in the error message, the OPed code is essentially the same as the documented examples. I don't understand the point you want to make.

Also on your while loop you have while (my $line = <>){ without the file handler.

I think that Peter Keystrokes wants to specify the input file on the command line, e.g.
    perl script.pl input.file
IOW, to read and write to separate files rather than appending to the input file as your final code example does | both your code examples do.

while (<$fh>) {
     chomp;
     if ($_ =~ /^>hsa/) {
     $count .= length $_;
         $count .= " ";
     }
 }
[download]

Update:

for (@lines) { print $fh length($_) . " " if ($_ =~ /^>hsa/); }
[download]

This loop counts | These loops count the characters in the lines beginning  '>hsa' rather than in the single string of the  'ACTG...' line that follows. The counts of 49 and 54 characters, respectively, for the two example data records reflect this. (And again, these counts are appended to the end of the input file rather than written to a separate file.)

Please believe that these comments are not meant to "pile on," but are only offered in the hope of smoothing the way a bit for future Seekers of Perl Wisdom.

---

Give a man a fish:  <%-{-{-{-<

Hello AnomalousMonk,

No worries no problems, I am always open to read comments from monks which have greater knowledge.

I am just commending to avoid confusion, OP on the open function has write mode > instead of read mode <.

Regarding:

I think that Peter Keystrokes wants to specify the input file on the command line, e.g.

OP defined a file my $filename = 'counts.txt';, so I assumed that he is defining the file through the script not through the command line. But never the less I was not aware of Using the magic ARGV handle - <> in perlop. as anonymous point out. Thanks for the tip. :)

Regarding:

This loop counts the characters in the lines beginning '>hsa' rather than in the single string of the 'ACTG...' line that follows. The counts of 49 and 54 characters, respectively, for the two example data records reflect this. (And again, these counts are appended to the end of the input file rather than written to a separate file.)

I know some how that my results is wrong but I still can not understand why I can not link the lines that start with >hsa and why it only captures the ATG.... Can you provide me a bit more information? (Thanks in advance for your time and effort)

Seeking for Perl wisdom...on the process of learning...not there...yet!

Thank you guys I am trying to absorb all your feedback (back and forth) as I'm pretty much a novice at this. I have been trawling the web taking bits of code and trying to make something work albeit not perfectly. At least here I can try to refine what little I have learnt. I appreciate this.

I am wondering how you are reading the file.

Using the magic ARGV handle - <> in perlop.

I do not know exactly why you want to read and write to your file

OP never said they wanted to.

Your code produces different answers from the OP's, the correct answer is "107 251 ".

for(@lines) has no advantage over while (<>) here, it's slower and takes more memory.

Hello Anonymous Monk,

I had no clue about that, the only reason that I would use like this is to save some coding lines.

But if while (<>) is faster then the obvious solution would be this.

Thanks for providing this information.

Seeking for Perl wisdom...on the process of learning...not there...yet!