wordcount can help in this

$zcat test.fq.gz | head -5000 | seqtk fq2fa - >test.fa

$ wordcount test.fa -wordsize=9

output:

$ head test.wordcount
AACAACAAC    44
CAACAACAA    44
ACAACAACA    44
TCACTCACT    20
CACTCACTC    19
TAGGGTTAG    17
TTAGGGTTA    17
CTCACTCAC    17
GGGTTAGGG    15
GTTAGGGTT    15

try this one here. Haven't tried it myself yet, but this is something I also need and will try it soon.

This has already been answered here C: script to calculate length distribution of a collapsed fasta file

The input you have is fastq which is different from fasta. so modify the solution accordingly.

I presume you want to count the frequency of the read you specified in your fastq files.

Here is a little perl script (script.pl) you can take use:

#!/usr/bin/perl -w
use strict;
# if your fastq file is gzip compressed, you may need to add PerlIO::gzip module
use PerlIO::gzip;

# usage: perl script.pl <specified_seq> <R1.fastq R2.fastq | R1.fastq.gz R2.fastq.gz>

my $seq = $ARGV[0];
my $count = 0;

foreach my $i (@ARGV[1..$#ARGV]){
    open IN,"<:gzip",$i || open IN,$i || die $!;
    while(<IN>){
        # read in four lines of each read in fastq format file.
        <IN>;
        my $read = <IN>;
        <IN>;<IN>;
        if ($read eq $seq){
            $count++;
        }
    }
    close IN;
}

# output counts
print "$seq occurs $count times\n";