Opinion needed on Truseq vs Scriptseq

[hoytpr]

So quite simply, when doing the math, 48 bacterial RNA-seq samples (libraries) would cost over a thousand bucks more using Truseq + Ribozero, versus using ScriptSeq "complete" + Scriptseq "barcodes" + Lucigen's Failsafe PCR mix. (for this small core Illumina discounts don't even cover shipping).

This amount of money is significant to the researcher. But we haven't used the ScriptSeq and already have all the DUI's for the TruSeq kits.

Can I get a few opinions about TruSeq vs ScriptSeq? Could we safely use the ScriptSeq kits? Even if one does everything correctly, will these kits cluster about the same?

Thanks
-pete

[JakobHedegaard]

We have been using Ribo-Zero + ScriptSeq for years with success. Primarily for samples of human RNA (using Ribo-Zero GOLD or Globin-GOLD) but also a few samples of bacterial RNA (of course using the Ribo-Zero bacteria). I havnt tried any of the TruSeq RNA kits, but know that the method for strand-specificity differ, TruSeq is based on dUTP while ScriptSeq use tagged cDNA synthesis.
A trick to save even more: The Ribo-Zero kits comes in high or low format for the same price, but the high kit contain app 2x reagent. So we purchases the complete kits in the high version and additional ScriptSeq kits to supply - and follow the low input protocol. Using 500 ng total RNA as input we can hence get two Ribo-Zero reactions out out each High Ribo-Zero reaction.
If you are going to use all 48 ScriptSeq indexes, be aware that index 11 and 31 only have two mismatches to index 41. So use demultiplexing with no-mismatch option.
Best, /Jakob

[hoytpr]

Thanks for the great advice Jakob.

Pete

[UCan'tBcereus]

Hi Pete,

Never used the Truseq Kit before, but our lab uses Scriptseq Complete Gold Epidemiology Kit + RiboZero when we make RNASeq Libraries. We use epidemiology kit since we are often working on metagenomic samples; the epidemiology kit removes rRNA for both the bacteria and the host.

Getting high quality Total RNA is difficult with metagenomic samples, but we have successfully made libraries/sequenced of lower quality RNA. Basically, if I see a 16S and 23S peak when QC'ing the total RNA, I can make a successful RNASeq Library. So if RNA Quality is a concern, I wouldn't worry about when using the Scriptseq Complete Gold Epidemiology Kit + RiboZero. High quality RNA works fine too. Our libraries cluster just fine on either the NExtSeq or the Hiseq 2500/4000 when we sequence them.

Originally Posted by JakobHedegaard

...
A trick to save even more: The Ribo-Zero kits comes in high or low format for the same price, but the high kit contain app 2x reagent. So we purchases the complete kits in the high version and additional ScriptSeq kits to supply - and follow the low input protocol. Using 500 ng total RNA as input we can hence get two Ribo-Zero reactions out out each High Ribo-Zero reaction.
If you are going to use all 48 ScriptSeq indexes, be aware that index 11 and 31 only have two mismatches to index 41. So use demultiplexing with no-mismatch option.
Best, /Jakob

I completely agree with the above. Good points to keep in mind. Both the "Low Input" and "Regular Input" protocols have worked equally as well for us. To alleviate that indexing issue, you could just make 4 smaller 12 sample pools, instead of making a big 48 sample pool and sequencing across multiple lanes. We usually sequence a 12 sample pool on one Hiseq Lane.


Let me know if you have any other questions.

[hoytpr]

UCan'tBcereus: Your post could not have been better timed. We have a client with some very precious samples that show both peaks, but the RIN scores are very low. We were considering turning down the sequencing request, but if what you say is true, we might be able to pull this off.
Thanks very much!

-p

[UCan'tBcereus]

Originally Posted by hoytpr

UCan'tBcereus: Your post could not have been better timed. We have a client with some very precious samples that show both peaks, but the RIN scores are very low. We were considering turning down the sequencing request, but if what you say is true, we might be able to pull this off.
Thanks very much!

-p

Hi Pete,

Glad to be of help. Also, just to be clear, when I say successful library prep, I mean that the Library looks good and sequenced/clustered well. As far as actual quality of data, I cannot be sure, since I do not analyze the data myself. But there were no complaints from the labs where we made libraries with poor quality RNA that we worked on.

As far as RIN scores, sometimes the actual number could be off. The RIN seems to be less accurate since we switched to the Tape station. Having both of the peaks was a better indicator of library success, rather than a hard number of something like >3 RIN.

Feel free to ask any more questions!

[hoytpr]

Right, I suspect the final data might be affected, but in this case, the project was probably at least a year old. We just want to help.

We have both a Tapestation and 2100 and they tend to agree with each other, but after several decades, I can usually just "look" at an RNA and tell if it'll sequence regardless of the RIN score. The spec readouts are good for detecting contaminants though.

Thanks again.