This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Why are GC% per base important in quality control of reads?

Hello,

In quality control of reads, why do we look at the GC% per base position? I have the following result

Gem. lengtes: 75
Max. lengte: 101
Min. lengte: 24
GC globaal: 32%
GC per base position: 
[32, 33, 33, 33, 33, 33, 33, 32, 33, 33, 33, 33, 33, 33, 33, 32, 32, 32, 33, 33, 32, 33, 33, 33, 33, 32, 32, 32, 32, 32, 31, 31, 32, 31, 31, 31, 31, 30, 30, 30, 30, 30, 30, 30, 30, 30, 29, 29, 28, 28, 28, 28, 28, 27, 27, 27, 27, 27, 27, 26, 26, 25, 25, 25, 25, 24, 24, 24, 23, 23, 22, 22, 21, 21, 21, 20, 19, 19, 18, 18, 17, 16, 15, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 1, 0, 0]

Looking at the GC per base position, I can see that the GC% per base position decreases. So what can I conclude from this? Thank you in advance!

qc assembly
• 2,650 views • link
updated by gb gb • written by c.clarido c.clarido

Looks like it is just biased because of the read length. After 24bp the number of read is decreasing, as the GC%.

• 1 views • link
written by corend corend

I also believe that there is a rule that you should mention that it is about a school assignment. Maybe a moderator can confirm that.

• 1 views • link
written by gb gb

1 answer

In your case it can be a poly-A tail or something, did you trim of the primers/adapters and everything after the primers/adapters? Or will this still look like this after quality trimming? Are all the reads the same length?

I think you mostly use GC-content as a quality check if you compare it with the GC-content of a reference. So you expect that a certain species or chromosoom has a certain "specific" GC%. If you expect 40% on chromosoom x and it is 75% something is off.

EDIT:

I just noticed that there is a big difference in shortest and longest read so that plays a roll

• 1 views • link
written by gb gb

Yeah they look like this after quality trimming, so the lengths varies a lot. So I can assume they could be poly-A tail?

• 1 views • link
written by c.clarido c.clarido

You could scan/trim for polyA and see if the length reduces even further.

• 1 views • link
written by GenoMax GenoMax

No you can not assume they could be poly-A tail. That's something you can see in de sequences. But if it are not poly-A tails it does not mean your data is wrong. You just have this result because the differences in length.

• 1 views • link
written by gb gb

Log in to answer this question.

More posts like this