I generally analyze paired-end illumina platform geretaed fastq having 6mer barcode. For ex,
@K00171:40:H3KJTBBXX:8:1101:20518:2020 2:N:0:CGATGT
GCTGCTTCAATCCTTTCCTCTACTGTCATGTCACTTCGGGCTTTCCCCTCTTTGACATCAATAAGTTCAACTTCATAAACAAGGTCTGCCATCAATCGTTA
+
A,A,F<KAFKFKFKFKKFKAFAFKAKKKFKKAFKKKKKKKKAAAAKKKKKKAKKFKKKKKKKFAKFA7FFFKKKKKFKFKFKFKFKKKKKKFFKKFKKKFK
For removing the adapters, I run the following trim_galore command,
trim_galore –paired –phred333 –q 20 –a GATCGGAAGAGCACACGTCTGAACTCCAGTCAC[6BASE]ATCTCGTATGCCGTCTTCTGCTTG – stringency 5 –e 0.1 –t –r1 35 –r2 35 pair_1.fq pair_2.fq
But now I got fastq files from sequencing company with 8mer+8mer barcode. For ex header is like ,
@K00171:711:HVYHKBBXX:4:1101:28036:1332 1:N:0:ACACAAAA+GGAGATCA
I couldn't find any documents regarding this system in the pdfs. Can anyone suggest my how the adapter trimming can be used in this case?
The sequencing protocol,
Library Kit : TruSeq RNA Sample Prep Kit v2
Library Protocol: TruSeq RNA Sample Preparation v2 Guide, Part # 15026495 Rev. F
Thanks @Devon for the suggestion.! BTW, I' not sure how the 8mer+8mer case come? If you could give a search keyword for this kind of barcoding, it would be helpful for me to further look into to available documents.
You have dual indices, that's pretty common.