How to fix mix-orientation reads

User Support
1

I’ve been searching the answers in Q2 forum, some clues but still have low confidence that how to fix this issue.

Hiseq 250bp data, paired-end. I think that there are forward and reverse reads in both R1 and R2. I think it would be best is to have the files in the same orientation in each file, but have no idea how to do that by Q2.

Example. Primer VH1_2 (AGGTCCAACTGCAGCAGCC) can be found starting at position 2 in R1 or R2.

One pair where it matches R1.

R1:

@SNL147:722:HGCMVBCXY:2:1106:18591:4974

CAGGTCCAACTGCAGCAGCCTGGGGCTGAGCTTGTGATGCCTGGGGCTTCAGTGAAGCTGTCCTGCAAGGCTTCTGGCTACACCTTCACCAGCTACTGGATGCACTGGGTGAATCAGAGGCCTGGACGAGGCCTTGAGTGGATTGGAAGGATTGATCCTAATAGTGGTGGTACTAA

GTACAATGAGAAGTTCAAGAGCAAGGCCACACTGACTGTAGACAAACCCGCCAGCACAGCCTACATGCAGCTCC

R2:

@SNL147:722:HGCMVBCXY:2:1106:18591:4974

GAGGAAACGGTGACCGTGGTCCCTGTGCCCCAGACATCGAAGTACCAAACCCCGTCCCAGTTTGATCTTGCACAATAATAGACCGCAGAGTCCTCAGATGTCAGGCTGCTGAGCTGCATGTAGGCTGGGCTGGAGGGGTTGTCTACAGTCAGTGTGGCCTTGCTCTTTGACTTCTC

ATTTTACTTAGTACCACCAATATTAGGATCAATCCATCCAATCCACTCAAGGCCTCGCCCAGGCCGCCGCTTCA

One pair where it matches R2:

R1:

@SNL147:722:HGCMVBCXY:2:1106:7141:4616

CAGGTTCAGGTGACCGTGGTCCCTGTGCCCCAGACATCGAAGTACCAGCTCTACTACCGTAGTCCCCTTGCACAATAATAGACCGCAGAGTCCTCAGATGTCAGGCTGCTGAGTTGCATGTAGGCTGTGCTGGAGGATTTGTCTGCAGTCAGTGTGGCCTTGCCCTTGAACTTCCC

ATTGTAGTTAGTATAACTATCAGAAGGATCAATCTCTCCAATCCACTCAAGGCCATGTCAAGGCCTCTGCTTTG

R2:

@SNL147:722:HGCMVBCXY:2:1106:7141:4616

CAGGTCCAACTGCAGCAGCCTGGGGCTGGACTTGTGAAGCCTGGGGCTTCAGTGAAGCTGTCCTGCAAGGCTTCTGGATACACCTTCACTAACTACTGGATAGGTTGGGCAAAGCAGAGGCCTGGACATGGCCTTGAGTGGATTGGAGAGATTGATCCTTCTGATAGTTATACTAA

CTACAATGGGAAGTTCAAGGGCAAGGCCACACTGACTGCAGACAAATCCTCCAGCACAGCCTACATGCAACGCC

2

Hey @mimisikai - we have added support for this in the upcoming QIIME 2 2020.6 release, currently scheduled for public release tomorrow. If you want to test it out (using a pre-release version) see the install instructions here:

https://dev.qiime2.org/latest/quickstart/

changelog:

3

Thanks, @thermokarst. Are you pointing cutadapt function? My data has been de-multiplexed. Is that also working so? And could you tell where I can find tutorial about that? Thanks.

4

Yes, sorry for not making that clear.

Ah, unfortunately this won't help you. Can you get your hands on the multiplexed reads?

There is no tutorial at present.

8

Hi @thermokarst, It seems rawest data they can provide. I wonder QIMME2 must handle sequence with multiplexed reads? I don’t want to cut any sequence in my read if they’re already demultiplexed. I only need to orientate them to make sure they’re all in the same direction. Because in the next step I need to annotate the sequence based on the For and Rev primers.

9

I’m pretty sure I saw a thread in the forum said R1+R2 --> new R1, and R2+ R1 --> new R2, then somehow input newR1 and new R2 into Q2 for processing should be fix out this problem. I try to find this post but I cannot.

11

In general, no - but in this case, if you had multiplexed reads, you could use the new mixed-orientation demux in q2-cutadapt 2020.6.

If you can get the multiplexed reads (or somehow make them by concatenating reads together), you could use q2-cutadapt.

13

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.