how to test number of 'AAAAA' segment in reads file?

[super0925]

Hi all
A naive question.

I want to test how many reads contain the segment of 'AAAAA'?

Is this command work?

grep 'AAAAA' -c sample1.fastq

Q1:
If it is OK. why the counts result is different from the statistic of 'AAAAA' on the Kmer content slides in FASTQC software ?
Q2:
How to how many reads contain the segment of >5A? ('A' continously more than 5 times sth like AAAAAAA,AAAAAAAAAA)
Q3:
If I want to count the reads not only contain 'AAAAA' but also with end of 'AAAAA'
I mean ACTG..........AAAAA.
How could I do it?

[holmrenser]

First off: your grep will 'work', as in that it will return some value.

Q1:
My guess would be that it is different because your grep only returns line numbers. So for a read containing the pattern twice it will only return a count of one.
Q2:
You could try a grep with a regular expression.
Something like: egrep -c '[A]{5,100}' sample1.fastq

5 = lower bound
100 = upper bound, assuming your reads are max 100bp

[super0925]

Thank you!

[GenoMax]

For Q1:

$ grep -oh string_being_searched file | wc -w

Will account for multiple occurrences on a single line and give the correct count.

[super0925]

For Q1:

$ grep -oh string_being_searched file | wc -w

Will account for multiple occurrences on a single line and give the correct count.

Thank you! I don't want the multiple counts in a single line . but what you post here is so useful and probably for my furture work.
PS:
If I want to count the reads not only contain 'AAAAA' but also with end of 'AAAAA'
I mean ACTG..........AAAAA.
How could I do it?

[m4merg]

egrep "AAAAA$" input > output
I also suppose that when you work with fastq line with quality can disturb information u need. So id better created fasta and worked with it

[super0925]

egrep "AAAAA$" input > output

Thanks!!!!