Fragmentation conditions - duration, temperature?

[Lutz Froenicke]

Hello All,

we did our first experiments with the Smart-3SEQ default protocol (including the recommended fragmentation conditions:  5 min at 80C)

We did start out with 90 ng high quality RNA (RIN score 9).

We got great yields, but our libraries reproducibly ended up longer than expected - with a peak at 600 bp. These are certainly still sequence-able.

Nevertheless all other Tag-Seq type protocols we are aware of, as well as all the tapestation traces posted here show fragment lengths skewed towards shorter fragments (peak at around 200 bp).

Should we extend the fragmentation time?

Has somebody else used the protocol for high-integrity RNA samples?

Thanks in advance!

[Joshua Tay]

Hi Lutz,

Could you attach your tapestation trace?

Thanks,

Josh

[Alejandro Pezzulo]

Hi,

We've used the default protocol for high quality RNA and our peaks are mostly around 600. These libraries have yielded very good alignable reads.

I'd be curious to hear of other's experiences.

Alejandro

[Joe Foley]

Yes, the protocol for intact RNA is meant to generate fragments that are unnecessarily long for tag-based sequencing, because they're easy to size-select away from potential adapter dimers, which are also longer than usual in this protocol - hence the different size selection conditions for intact vs. fragmented RNA in the protocol. Given those long inserts, you could sequence longer reads to increase the number of alignable positions in the transcriptome, but this has diminishing returns even in ideal conditions (see Supplemental Figure S30 in the paper) and is probably not worth the extra cost in most cases.

If you still want to recalibrate the fragmentation for your RNA samples, I would recommend varying the temperature rather than the time. Time is a little harder to control, depending on how quickly you load the thermal cycler and how quickly it ramps, and in many instruments these days you can actually set different temperatures in different parts of the heat block so the calibration experiment is very easy. My intuition is that the fragmentation is asymptotic anyway: the longer the fragment, the more likely it will break, so after some amount of time the apparent rate of fragmentation will be negligible. If so, rescaling the whole fragment size vs. time curve by changing the temperature may be easier than fine-tuning the timing in the steep part of the curve.

--
You received this message because you are subscribed to the Google Groups "Smart-3SEQ" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smart-3seq+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/smart-3seq/2c49ca47-273c-4fc4-839a-af5586d7d55d%40googlegroups.com.

[Lutz Froenicke]

Hello All,

thanks for the replies. This is an example trace.

It seems the long fragments are expected?

Thanks! 

[Lutz Froenicke]

I hope the image does not shrink this time.

[Joe Foley]

Yes, that looks like a good size distribution.

However, it might be a little overamplified: there's a long tail of molecules that migrate more slowly than the main peak, which appear as longer molecules but could actually be complex secondary structures that form as artifacts of too many PCR cycles. At this degree they don't cause major problems downstream, but you could probably cut it back by one or two cycles and still get a good yield from this amount of RNA.

On 10/18/19 5:24 PM, Lutz Froenicke wrote:

Hello All,

thanks for the replies. This is an example trace.

It seems the long fragments are expected?

Thanks! 

On Friday, October 18, 2019 at 11:24:39 AM UTC-7, Lutz Froenicke wrote:

Hello All,

we did our first experiments with the Smart-3SEQ default protocol (including the recommended fragmentation conditions:  5 min at 80C)

We did start out with 90 ng high quality RNA (RIN score 9).

We got great yields, but our libraries reproducibly ended up longer than expected - with a peak at 600 bp. These are certainly still sequence-able.

Nevertheless all other Tag-Seq type protocols we are aware of, as well as all the tapestation traces posted here show fragment lengths skewed towards shorter fragments (peak at around 200 bp).

Should we extend the fragmentation time?

Has somebody else used the protocol for high-integrity RNA samples?

Thanks in advance!

--

You received this message because you are subscribed to the Google Groups "Smart-3SEQ" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smart-3seq+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/smart-3seq/2f02e1cd-67ff-4d77-93ea-35524aab1b61%40googlegroups.com.

[Lutz Froenicke]

Yep, the yield was better than expected, likely because we started with almost perfect RNA samples