SMARTer Ultra low RNA Kit Primers are found in sequencing data and Tophat questions

[TEFA]

Dear All,

I am working with SMARTer ultra low RNA kit and TruSeq libraries for Illumina sequencing. I have sequenced my paired-end libraries in GAIIx (reads of 36 nucleotides) and some of them in Hiseq2000 (reads of 100 nucleotides).
I analyzed the quality of the RNA-seq data and I found overrepresented some sequences that correspond to the primers used for the Kit (that are 30bp long), this sequences are present in 25% of the total number of reads (for example one of them: AGCAGTGGTATCAACGCAGAGTACATGGGATGGCA).
Does somebody have a similar issue???.
I am planing to remove these sequences from the data (using the tools to remove adapters FASTQC or FASTX-toolkit), but for the RNAseq of 36bp I will loss all the reads that contents this sequecence. I will use Tophat to map the reads after remove the adaptor and I don't know If there is a problem to run this program when the fastq files from left and rigth (R1 and R2) doesn't have the same number of reads. Anyone have an idea???


TEFA

[weigrc]

Hello TEFA,

Did you do size selection by cutting gel to get rid of primer dimers?! Since this low DNA input may cause some primer dimers formation. Primer dimers should be 240 bp something.

We are going to use the SMARTer kit and following GAIIx sequencing as well. One point we discussed with Illumina is they recommends to apply PE sample prep kit instead of TruSeq DNA sample prep kit for the downstream DNA sample preparation because they've validated the PE kit, but not the TruSeq one, is compatible with the DNA input lower to some 10 ng DNA.

May I ask you a couple of questions?

Did you evaluate the output after the SMARTer kit?!
and how much the total RNA input you used for the SMARTer kit?!


Thanks,
Wei

[Conblauw]

Dear all,

Any news about the smarter sequencing kit or any experiences that someone wants to share? I just started using it and would like to hear what other peoples experiences are...

Thanks in advance!

[neurohugo]

Question about SMARTer kit

Hi all!
I recently tried SMARTer and Nugen RNAseq V2 kit and it seems that my SMARTer kit did not synthesized the very last portion of cDNA(where primer binds) and I did get nothing.

I tested with other amplification kit which works (tested) and qPCR to test which process could have gone wrong... It seems that, from qPCR with genes which express highly (Vimintin), cDNA is definitely synthetized.

Is there anyone have the similar issue?

Thanks

[vanillasky]

Check out these repeats

I'm seeing a lot of repeats showing up in my sequences as well. See file attached for a glimpse of the repeats after trimming and rRNA removal. I have filtered out all rRNA reads from this file. I also posted what the final dscDNA looked coming out of the Low Input RNA kit in another thread in this forum.

Did you observe a similar result (i.e. similar banding patterns)? For this low input you can't just run a gel and cut things out. You have to run samples with a High Sensitivity and Pico Chip on the Bionanalyzer to assess if the depletion and cDNA synthesis steps worked with the low input rRNA. Hope we can help each other out.

[SS00]

I actually had a similar issue and contacted Clontech about it and they didn't really seem surprised and their recommendation was to filter out the adapters. When I brought up the point that I would be losing read length they suggested increasing read length in anticipation of there being adapter sequences.

It seems strange because in the SMARTer Universal kit they actually remove the adapters with RSaI I think, but apparently the Illumina chemistry doesn't allow for them to do the same.

[vanillasky]

So are you saying that most of the adaptor sequences are coming from the illumina chemistry or from the Smarter kit?

[vanillasky]

Also, what illumina kit did you use? We used a NEB kit.

[magnolia93]

As a scientist at Clontech, I will let you know that we understand this phenomenon. It is explained in the new version of the SMARTer Ultra Low user manual. But briefly, the adapter should only be present in your read 2 (from a paired end read) and will only be present if you are using a ligation-based adapter addition.

You can do an RsaI digestion, if you would like to remove the adapters (it works), but it will maintain the dT30 that is used for the original priming. The presence of this stretch we find is more disruptive to people (both because it uses a lot of cycles and because it is a long stretch of low complexity which can affect the base calling).

[SS00]

vanillasky, I actually used the Truseq kit but as magnolia93 helpfully explained (thanks!) it is just the fact that you need ligation based adapters for Illumina Sequencing.

magnolia93, when I did have these overrepresented sequences I did Paired End reads. Going forward I intend to do single reads which you said shouldn't be a problem. Has this been tested? That is good news!

[magnolia93]

SS00, yes it has been confirmed. You can see it clearly by the base distribution right off the instrument (we show a figure of it in our user manual).

But, to be clear, ligation based illumina adapter addition is only present in some kits (TruSeq, NEB), but the Low Input Library prep kit that Clontech sells uses a modified ligation that does not lead to this phenomenon. Also, Nextera and Nextera XT do not have this phenomenon, and they are also a little better at capturing the 5' end of your transcript than the other library prep methods.

[Wonghe]

low mapping rate

Hi, I constructed a RNA-seq library using the clontech Smarter total stranded RNA low input kit. However after mapping using BWA and trimming the adapters I only manage to get 50% mapping rate.
I'm wondering if there is a need to trim off the 1st 6 bases of the random primers before mapping?
Or what else can be done to improve the mapping?

[efrainrib]

Same question as Wonghe, also only mapping ~50% of reads with the first 6 bp of read 2 always failing the per-base seq quality badly. Waiting on an answer from clontech about this but it seems to make perfect sense that this is the Stranded N6 Primer (Yes-it is 6bp long!) and that it should be trimmed. not sure if it will help the mapping but will report back once i do it