How to perform this experiment.

[Ivan Castro]

Hi to all. Unfortunately I still without understand completely the NGS technology. So, I need that someone give me some tips or workflow to carry out succesfully this experiment in a Miseq machine:

I want to sequence the 5S rDNA (~1500bp), the 18S rDNA (~1700bp) and the mitochondrial genome of 140 individuals of elasmobranchs (sharks and rays), so I will execute more than 420 sequencing. The mitochondrial genome will be amplified by the Multiple Displacement Amplification technique (http://en.wikipedia.org/wiki/Multipl..._amplification).

I read that I need the Nextera XT kit and the guide seems to be easy, but I not understand yet (if is neccesary) how to use the dual indexing to sequence my samples.

I appreciate any comments to help me clarify this problem.

[ECO]

Dual indexing only allows you to multiplex more samples. Sounds like you need to do a lot more reading, and perhaps some in silico experimentation to understand exactly how much coverage you need for accurate mapping.

The calculation will look something like:

(total target size per individual) *
(coverage of each target desired) *
(some fudge factor for sample prep biases, I'd use >5x)
== some number of bases.

Divide the specs on your favorite configuration of Illumina machine by your number and that will tell you how many individuals you can load per flowcell/lane (and thus how many unique barcodes you'll need in the experiment).

Good luck.

[akorobeynikov]

Another thing to consider: if you will want to assemble your data de novo instead of mapping, then you will need to use an assembler which can cope with huge variations of coverage and MDA-induced chimaeras. SPAdes and IDBA-UD are tools of the choice here.

[Ivan Castro]

Hi, thanks to all for reply. @ECO, that info is new for me and you must take into account that I did never touch a sequencing machine and that I will never touch one until the moment of make the sequencing (my situation is not easy). All that I need to know for make the sequences is only by read manuals, there is not someone that can explain me in person how to use the Miseq machine.

@akorobeynikov, thanks for the info. Really I was thinking about that problem because the 5S rDNA is highly (very) variable, even among species of the same genera. So a manifest file with a sequence of reference is impossible for the experiment.

Another doubt that I have is if possible to mix the three amplicons for the same individual (5S from n1, 18S from n1, and mitchondrial from n1) and handle the mixture as one sample (sample n1) and apply the Nextera XT kit for this sample (n1) and for the other 140 individuals (n2, n3, ..., n140). Is possible?

[microgirl123]

As far as actually using the MiSeq, you just stick in a flow cell, poke a hole in the reagent cartridge, stick in your sample, toss the reagent cartridge into the machine, and hit go. So all you need to know about using it is easily found online!

It's the library creation that is the hard part! First you will need to decide how many MiSeq runs you want to do (and can afford to do). Illumina has a coverage calculator (http://support.illumina.com/download...alculator.ilmn) that will help you decide.

Once you know how many runs you are going to do, the indexing comes in. Indexing is what will allow you to tell your samples apart if you run more than one together on the flow cell. Dual indexing (having an index at each end of your sequence) allows you to multiplex more samples than single indexing (an index at only one end of the sequence). Nextera XT automatically uses dual indexing.

[Ivan Castro]

Hey, Tx @microgirl123 for the tool.

Ok, reading more guides, seem to be that I will need the PhiX kit. However, the PhiX's manual state that it is not applicable if I will use the Nextera XT kit. This is confused.

Now that I understand how the dual indexing works, I have a question: with dual indexing I can run 384 samples together, but the amount of reagents (from each kit) reaches only 96 samples. So, if a want run 384 sequences I'll need to buy 4 kit and order the 40 primers anyway?

Thanks in advance to all.

[microgirl123]

I'm pretty sure that the Nextera XT dual indexing only will multiplex 96 samples - the 384-sample index kit (FC-131-1002 (96 indices, 384 samples)) provides enough volume of indices to make 4 sets of 96 indexed samples. You could not run all 384 samples together on one flow cell.

phiX is a ready-made library from Illumina that you can spike in with your sample when you load it onto the MiSeq. In cases where your sequences are all very similar it adds diversity to the reads. In other cases it just provides a control so that if your run looks strange, you can try to pinpoint the problem to the run itself or to your library.