Sequencing PCR amplicons on MiSeq

[rjlebbink]

Hi,

I'd love to receive some advice on a strategy to sequence PCR amplicons on a MiSeq.

I'd like to simultaneously sequence ±300 PCR products from control and 'experimental' samples on a MiSeq. The PCR products will be ±200-400bp long.
For this, I'll design ±300 primerpairs to amplify the regions of interest. Then, I'd like to sequence these simultaneously. What would be the best way to go

1) Perform 'regular' PCR with gene-specific PCR primers and pool 'control' PCR products and 'experimental' products on two populations. Then ligate on a barcoded adapter for PE sequencing? Pool and sequence? What kit would work best for this?

2) Would it be easier to add a short linker to the primers and use these as priming sites in a next PCR reaction in which barcoded adapters are added to the pooled PCR products? If so, how long should this adapter be? And did anyone try this method before (what sequence would you recommend)?

How problematic is it when during sequencing all clusters give the same nt sequence (low complexity at the same region in the DNA). The entire chip may light up, does this make it hard to base-call? What are the solutions? I only have two pooled samples (control vs experimental), could I use a mix of barcodes for each? Or should I mix in the PhiX library? At what ration?

I'd prefer the first option, since it's less work. I'm just not sure how efficient the adapter annealing is in the kits.

Any help/advice would be much appreciated.

cheers

[microgirl123]

I'm trying to get my head around what you are trying to do!

You only have two samples, but will generate 300 different PCR amplicons from each? So you'll end up with 600 different sets of amplicons that you will pool together relative to their sample. Then you want to add adapters and indices to each sample (2) and load them onto the MiSeq?

[GenoMax]

I think rjlebbink means two categories of (not just two samples).

[BBthekid007]

We've found it easier to do the latter of your two suggested processes. For the 'linker' (also called a universal tag), we use the Illumina sequencing primer (but not the complete adapter). For the second round, we use essentially a full TruSeq adapter, including the index and sequencing primer. A setup like this, in which all of your amplicons have the same universal tag, would mean you could pool all of the control and experimental amplicons into a single 2nd round of PCR. That wold make the workflow as simple as your ligation method.

For very low diversity libraries, we typically design multiple primers for the first round of amplification that contain degenerate 'offsets' to stagger the resulting sequence and create more diversity at each position. It would look something like this for your first round forward primers:

FwdPrimer1: IlluminaSequencingPrimer --- GeneSpecificPrimer
FwdPrimer2: IlluminaSequencingPrimer --- NN --- GeneSpecificPrimer
FwdPrimer3: IlluminaSequencingPrimer --- NNNN --- GeneSpecificPrimer

Our experience has been that adding offsets in multiples of two produces the best outcome. Not sure why, but it may be that offsetting by only a single position causes problems with phasing/prephasing. Since phasing is only done one position ahead/behind, having two adjacent clusters offset by two positions should be less troublesome.

[rjlebbink]

Hey,

Thanks for the help here. It's indeed ±150 different PCR reactions to PCR defined regions of the human genome in a control sample and an experimental sample. 300 products in total, and I'd like to study the complexity at these ±150 sites (mutations etc) in the experimental sample vs the control sample.

Thanks BBthekid007. This is very helpful. I have a few question on this

1) What is the sequence of this Illumina sequencing primer?

2) Would the first round primers would require the index as well? Hence:
Fw_1: IlluminaSequencingPrimer -- Index -- GeneSpecificPrimer
Fw_1: IlluminaSequencingPrimer -- NN -- Index -- GeneSpecificPrimer etc?
If so, I'd need to design 2 Fw and 2 Rev primers for each amplicon, one for each index?

3) Adding the universal tag to each forward and reverse primer would require quite a big investment. Each primer would have to be extended to attach the seq primer/universal tag (±15-20nt?) + Index + offset). This will easily be an extra 25-30 nt/oligo (2.50-3 euro) times 300 primers (150 amplicons): 600 euro. Would there be a more cost-effective approach? Perhaps a smaller adapter that still works OK?

Cheers

[vrayon]

Hi,

I'm designing my amplicon-seq experiment for miSeq. I found the answers provided here very useful. However, I still have a few questions.

Is there a good strategy for pooling PCR fragments?

Also, I only have 24 amplicons for sequencing, do you think I can compensate for the low complexity by adding PhiX control (say 10-40%), instead of adding degenerate 'offsets', as suggested by BBthekid007?

How much of the illumina adapter did you incorporate in the primers for the first round of PCR?

Thanks!

[FlausFSU]

If you are just doing targeted resequencing of human genes, you can use something like the QIAGEN GeneRead Custom sequencing panels which provides 150bp amplicons in highly multiplexed PCR reactions. Most "panels" provide thousands of multiplex PCR primer sets of overlapping design in 4 tubes. You simply PCR amplify, pool the 4 different reactions together then proceed through library prep.

Unless you are working in a novel species, no need to re-invent the wheel and design your own primers.

(Full disclosure, I work for QIAGEN. Same holds true for panels from other vendors though)

[nucacidhunter]

Hi,

I'm designing my amplicon-seq experiment for miSeq. I found the answers provided here very useful. However, I still have a few questions.

Is there a good strategy for pooling PCR fragments?

Also, I only have 24 amplicons for sequencing, do you think I can compensate for the low complexity by adding PhiX control (say 10-40%), instead of adding degenerate 'offsets', as suggested by BBthekid007?

How much of the illumina adapter did you incorporate in the primers for the first round of PCR?

Thanks!

24 amplicons will provide enough diversity if they are not targeting orthologous regions. You can simply amplify target regions with target specific primers in separate well, clean, pool in required proportion and ligate adapters for library prep. For more than one sample, adapters with different index can be used to identify sample of origin.

[jtotheulie]

This is a very informative thread. I do have one question.

I am pooling 3 low diversity amplicons from 170 samples.

How would I cost effectively lib prep the samples with an illumina adapter and dual indexes?

I am going to run it on the MiSeq V3 with either a phix spike or random shot gun library.

[nucacidhunter]

The easiest way would be following Illumina's two step 16S protocol with dual index for each sample.

[jtotheulie]

Thank you. One more question.

If I were to order the 16s primers for the illumina two step protocol with the overhang could I use them for different sequencing machines.

For example, use the 16s primers with the overhang and do blunt end repair and run them through the oxford nanopore library prep protocol?

I am only going to be using illumina for confirmation sparingly and would like to have a good primer set on hand that would work with both.

[fanli]

I imagine that would work fine as long as you treat them appropriately afterwards

[nucacidhunter]

Thank you. One more question.

If I were to order the 16s primers for the illumina two step protocol with the overhang could I use them for different sequencing machines.

For example, use the 16s primers with the overhang and do blunt end repair and run them through the oxford nanopore library prep protocol?

I am only going to be using illumina for confirmation sparingly and would like to have a good primer set on hand that would work with both.

Technically it should be fine, but I am not sure if the short amplicons will be suitable template for Nanopore sequencing.

[jtotheulie]

Yeah this can be an issue with oxford but metagenomic sequencing has been a success recently.

Has anoyone modified the 16s illumina protocol for other amplicons. I want to use it for 16s, its, and 18s. If we can do this then we can use the same sequencing primers and sequence them in the same run. Avoiding the conserved sequencing issue that illumina has and the confusion of the EMB protocol that uses different sequencing primers between the 16 and 18s.

Any thoughts?

[nucacidhunter]

Yes and Illimina protocol also includes instructions for using 16S approach for other amplicons as well. Pooling amplicons from different regions for sequencing provides enough diversity to decrease PhiX spike in to 1%.