Read coverage and accurate sample prep

[Meyana]

Hello!

I used to be in a lab that did a lot of sequencing - however, all sequencing and prepping fell on few people in the lab. Therefore I have some overview of the process of sample prepping, but need help with making sure everything is properly planned for NGS work in my new lab. My boss is asking me to calculate costs, so we can plan what is most suitable for our budget.

Material: Mitochondrial DNA (may contain miniscule traces of nDNA)
DNA amount for prepping: 0.5-1.5ng (expected)
Planned prep kit: NEBNext® Ultra™ II DNA Library Prep Kit for Illumina®
(as it seems to be able to handle the very low input material that I will have)

Question 1
Since mtDNA is rather small (<20kb), you generally don't need too many reads pr. sample. However, I will be doing assessment of heteroplasmy, so I will need a good read coverage, so not under something like 3000x.

Can somebody help me with how to calculate how many reads I will need?


Question 2
Choosing indexes.
I will be running the samples on either a MiSeq or HiSeq, depending on how many samples we end up with/how many samples I can put on the MiSeq (this is a cost issue).
Will *all* indexes be equally good on either the MiSeq or HiSeq?
I understand that the MiSeq uses dual indexing as a standard, why I was thinking to buy these indexes: NEBNext® Ultra™ II DNA Library Prep Kit for Illumina®
However, I am very confused about choosing the proper indexes, and would like some input.


Thank you so much for any help you can provide

[Genetic Librarian]

Hi,

Question 1: Your output requirement is 16kb (target size) * 3000 (coverage)

Thus, per sample you would need 48 Megabases as output.
A MiSeq v2 2x 150 bp PE run e.g. has an output of 5 Gigabases, so theoretically you could pool 104 samples on one Run. Substract a bit of duplicates, unassigned samples, PhiX etc. and you could aim for ~80 samples / Run.
Basically, that is what the Illumina Coverage Calculator does .

You can calculate the number of samples on a HiSeq lane (Output 90 Gb / Lane) yourself (hint: this is absolute overkill vor mtGenomes).

Question 2: For mtGenome, we used the Nextera XT Kit with its associated Indices.

[Meyana]

Perfect, thank you so much!
I tried to use the Coverage Calculator, but I am not sure I really understood the results it gave me - so I figured it better to ask for assistance

Do you have any experience in starting with under the recommended 1ng DNA for the Nextera XT kit? My starting material is a small brain tissue sample, so I am worried about how much material will be available for me. I will test next week when all my reagents are delivered, but I really doubt I will reach 1ng mtDNA...

[Genetic Librarian]

We always amplified the mtGenome in two long range PCRs, so we were never limited in material.
At the end, our final libraries were quite large (fragment size) and output was good, so I guess it might work well with 0,5 ng as well.

[Meyana]

Thanks.
I really would like to avoid doing PCR prior to sample prep, as the mutations I will be investigating are probably at a relatively low frequency + there might be smaller/larger deleted regions, and any PCR induced mutations or PCR bias towards some fragments due to deletions will likely interfere with downstream analysis...

[torben]

Perfect, thank you so much!
Do you have any experience in starting with under the recommended 1ng DNA for the Nextera XT kit? My starting material is a small brain tissue sample, so I am worried about how much material will be available for me. I will test next week when all my reagents are delivered, but I really doubt I will reach 1ng mtDNA...

The easiest way to scale down is to reduce all the volumes in the reaction. I usually use half the volumes given in Illumina's protocol, and I have collaborators who only use a quarter of the volume. You need to have your DNA in a pretty small volume though (≤1.25-2.5 µl). I elute my DNA in pure water and use a speed-vac to reduce the volume, but be carefull not to overdry the samples. Alternatively, I guess you can also use the normal volumes and only reduce the Tagment Mix to matvh your DNA input but I've never tried this.

[Meyana]

The easiest way to scale down is to reduce all the volumes in the reaction. I usually use half the volumes given in Illumina's protocol, and I have collaborators who only use a quarter of the volume. You need to have your DNA in a pretty small volume though (≤1.25-2.5 µl). I elute my DNA in pure water and use a speed-vac to reduce the volume, but be carefull not to overdry the samples. Alternatively, I guess you can also use the normal volumes and only reduce the Tagment Mix to matvh your DNA input but I've never tried this.

Thanks for this very useful info! I was also thinking of cutting everything in half, but was unsure as to whether this would be a bad choice somewhere in the procedure. Nice to hear other people are doing this successfully!
Now currently awaiting a new Qubit, so I can accurately measure my samples, not trusting the NanoDrop for these small concentrations...

I will ask around for the presence of a speed-vac, will probably need one no matter how I do this, as I am eluting my DNA in a too high volume from my Ampure beads (anyone know how little elution volume you can get away with using pr. vol Ampure?)