multiplexing smallRNA

[moguin]

Has anyone multiplexed for smallRNA libray prep?

[Chen Rui]

Small RNA sample pre

Up. I'm also wondering.

[link1]

multiplex small RNAs

We are multiplexing an average of 8 microRNAs per lane using this strategy :
http://www.biooscientific.com/Detail...tGenSequencing

expensive at first, but very cost effective after you begin multiplexing 10-70 samples per flow cell.

[suludana]

Hi, very interesting the link.
Could you please tell me if you need to do a second read for detect the index or with a single read run it is enough?

[moguin]

multiplex smallRNA

It does not matter if the barcode is on the 3' or 5' adapter because the product is only 22-30bp. You will sequence into your 3'adapter near the end of the run. No need for PE. You do have to make sure the barcode is on the right end of the adapter.

[link1]

in terms of ligation based barcode bias it does matter where the barcode is. Adding the barcode to the 5' adapter is not a good idea because the RNA ligase 1 enzyme at this end will prefer some bases over others. Adding the barcode on the 3'adapter is preferred as there is very little to no bias. I recommend using a RNA Ligase 2, truncated version that has been modified by mutation to avoid sequence preference.

[scooter]

I have to politely disagree with Link1. I have observed quite a lot of bias with 3' adaptor ligation using any of the available RNA ligases; including using pre-adenylated adaptors. In my experience, if you alter the sequence of the 3' adaptor (especially at the 5' end) you will get a different profile and this is true using any RNA ligase.

I am personally suspicious of 'in-line' barcoding for small RNA as I have yet to see real data where someone has used exactly the same RNA sample with multiple barcodes and recover the same profile of miRNAs captured in the library.

If someone has this sort of data, I would love to see it!

I think adding the barcode via PCR and performing a second read is a much less biased method and while it may add some cost to the run this cost is relatively insignificant when you can simultaneously profile many samples at once.

The other weakness of in-line barcoding is that not all small RNAs are only ~22 nucleotides. There are a lot of small RNAs that are longer (>30 nt) and these will be in your sample, so in these cases you will likely be unable to read the barcode and these clusters will essentially be omitted from data analysis.

[RNAseqer]

mirna barcoding

We just performed a huge investigation into the barcoding of small RNAs on the 3'adapter and these are our results:
By intelligently designing the barcode sequence so that there is even spacing between the nucleotides, we did not observe any obvious bias when using 7 barcodes for the same sample. In essence we were able to run one flow cell with 7 samples and got identical miRNA profiles when we ran 7 samples in a single lane. I agree with the previous poster about the obvious bias on the 5'adapter. The 3'adapter barcoding has been very viable for us and we have succeeded on producing several hundreds of miRNA preparations this way. We are publishing this data, but would be happy to share these sequences by private message.