Illumina TruSeq index

[Greenleaf]

I’m going to prepare a sequencing library based on Illumina TruSeq. The protocol my experiment is based on uses single-end sequencing, but a colleague suggested that I should use the paired-end method, as it makes aligning the reads easier. Does this somehow affect the primer design and/or the indexing? I'm especially wondering if I have to use dual indexing in this case.

A colleague also suggested that I could design my own indexes in order to multiplex even more samples on the same lane when sequencing than you could while using adapters from a kit. Is it complicated to design good and functional indexes, or is it more safe to use the already published ones?

I hope these aren't completely stupid questions - and if they are, please have mercy on a poor PhD student who is just getting started with the wonderful and complex world of NGS.

Thank you already in advance for any replies, helpful links and/or suggestions!

[GSviral]

Hi Greenleaf,

I typically always perform paired-end sequencing to maximise the data output from whichever sequencer I am using. In terms of alignment, I would hazard a guess that paired-end sequencing would lead to an easier alignment as the aligner would have a greater accuracy with double the reads if you get what I am saying (paired read 1 and read 2 as opposed to just read 1).

In terms of indexing, single-end and paired-end is same approach. As you've mentioned there are 'single' indexes and dual indexes. The main advantage of dual indexing is that you can prepare and sequence a greater number of samples in one run when compared to single indexing.

I would say as a Ph.D student, your best bet would to just stick to currently available indexes. I was in that boat myself and you would have enough to worry about without troubleshooting custom designed indexes :P

So really it all depends on your application. Which TruSeq kit are you using? How many samples are you multiplexing in one run? Which sequencer are you using? That would help give a more solid answer!

Hope this helps!

[Greenleaf]

Dear GSviral, thank you a lot for your reply and sorry for the delay!

I'm not using a kit but a previously published protocol with modified sequencing adaptors (which is one of the reasons for my troubles... A kit would be so much easier, but I guess I'll have to take this thorny path as a great learning experience.) Also, I'd like to multiplex 96 samples in a run and use the HiSeq2500 sequencer, but unfortunately Illumina's TruSeq 6mer indexes have only 48 sequences.
Currently I've been looking at the Meyer & Kircher 2010 (doi:10.1101/pdb.prot5448) and Bronner et al. 2014 (doi:10.1002/0471142905.hg1802s80) papers and trying to figure out if it'd be possible to use their indexes. If someone has any previous experience in using these, I'd love to hear if they worked! Any pointers on checking if the indexes are different enough to use would be also much appreciated

[nucacidhunter]

You can buy adapters from several vendors (Illumina, Bioo Scientific, Roche and others) with up to 384 index without buying a full kit.

[Greenleaf]

Thank you for the tip, nucacidhunter, I'll look into it in more detail. This far I've somehow found only systems where the adaptors and primers are mixed together (eg. the NEBNext system) and this is not a suitable approach for my current project.

[nucacidhunter]

Following lists some standalone forked (Y) Illumina style full-length adapters:

GeneRead Adapter set A and B (12 each) from Qiagen
NEXTflex® DNA Barcodes and NEXTflex-HT Barcodes (up to 384) from Bioo Scientific
KAPA single- and dual-indexed adapters from Roche
Integrated DNA Technologies various types including Unique Dual indexed with or without UMI up to 384
Illumina TruSeq and IDT for Illumina up to 96 indexes

[Greenleaf]

Thank you for the list, nucacidhunter!

I've now decided to change my approach a bit and go for the Illumina TruSeq CD dual indexes.

So far I've understood that I'll have to reverse-complement the i7 indexed primer from Illumina's list in order to get it bind the sequencing adaptor. Is this correct? Do you know if I have to reverse-complement the index sequences as well?

[nucacidhunter]

If you are going to use commercial full-length Y adapters or synthesize oligos to make up in house adapters then you do not need to change anything. The reversing P7 and i7 index is required if you are adding index by PCR to an adapter overhang.

[kwaraska]

Indexes

Illumina also has 96 unique dual indexes available for purchase that are both i5 and i7 in the same well.

The others are combinatorial and much more prone to mixup.