This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Adaptor trimming for RNAseq

I have a few samples of RNA-seq data. I have this adaptor sequience (AGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNATCTCGTATGCCGTCTTCTGCTTG (NNNNNN= 6 nt index)). Can someone please tell me what NNNNNNN =6 index indicates here? I was using BBMap tools to do the trimming and created the adaptor file like below, but I do not see any change in my new trimmed file (in terms of file size). 

>adaptor1
AGATCGGAAGAGCACACGTCTGAACTCCAGTCACNNNNNNATCTCGTATGCCGTCTTCTGCTTG
rna-seq
• 2,155 views • link
written by MAPK MAPK

firstly, file size is not always the best indicator of success or failure ;-) .

where did you get that adaptor sequence from? Did the seq facility provide that to you? I think they simply want to indicate there is a 6nt index present at the location of the Ns

• 0 views • link
written by lieven.sterck lieven.sterck

Yes, it was provided by the sequencing facility.

• 0 views • link
written by MAPK MAPK

and they did not provide any additional information, eg. what the index might be then?

long shot, but might they be present in the read headers in the fastq file?

• 0 views • link
written by lieven.sterck lieven.sterck

No, they did not provide anymore information, except for that adaptor sequence. The fastq header only has this info:

@K00317:102:HKVG3BBXX:4:1101:15889:1349 1:N:0:TGGTGAAG+TGGTGAAG
GTAGACTTGATAGTGATACCACGCTCTTGTTCAACGGCACGAGTATCGGGAGCTCTGGCATCACCAGCCTTTGCGGCGGA
+
AAFFFJJJ7FJFJJFJAJ<JJJJJFJJJJ<JJF-FFJJJJ-<JJJ7JJJ7FA<-A7FJJJFJJJAJAAJFAJ7AA7--<F
@K00317:102:HKVG3BBXX:4:1101:16295:1349 1:N:0:TGGTGAAG+TGGTGAAG
GGACAGGTCTGGCTCGGTTTTTGATATCTTCAGTTTAAGGGGGCGAATATCGTTACCCATAGATACCGAAAGTGGCCGAC
+
• 0 views • link
written by MAPK MAPK

What does the summary stats say about the trimming? Perhaps there are no adapters (always theoretically possible) or you were given an incorrect adapter sequence (also possible).

• 0 views • link
written by GenoMax GenoMax

1 answer

I would not worry about what NNNNN is, if the core sequence you are trying to remove is common. Use literal=AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC ktrim=r to get all variations of NNNNN.

• 0 views • link
written by GenoMax GenoMax

Thanks for your answer. So just to be clear, should I just have 

>adaptor1
AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC

and omit NNNNNNATCTCGTATGCCGTCTTCTGCTTG?

• 0 views • link
written by MAPK MAPK

Sure. As long as the leading sequence is identical for all adapters.

• 0 views • link
written by GenoMax GenoMax

or the trailing? [confused myself here]

• 0 views • link
written by lieven.sterck lieven.sterck

Once the first part is identified one generally wants to remove everything till the end on right so the trailing part gets automatically included.

• 0 views • link
written by GenoMax GenoMax

ok, so that was my confusion: the right side is the end thus (== not next to the actual sequence?)

• 0 views • link
written by lieven.sterck lieven.sterck

Log in to answer this question.

More posts like this