Inosine in B-adaptor of Rapid Library

[Zaag]

Example for RL13

A:
FAM/ CCATCTCATCCCTGCGTGTCTCCGACGACTAGACTCGACGT
Bold = key, followed by MID


B:

5Phos/CGTCGIGTCTIGTCGICG TCTCTCAAGGCACACAGGGGATAGG

Does anyone have an idea on why the inosines are used in the Y-adaptors?

[pmiguel]

My guess:

Y-adaptors of this sort would require an MID sequence that is complementary to the proximal end of the B adaptor.

Inosines should base pair promiscuously. So, by including those inosines, you allow the desired Y-adapter annealing to form, even across the MID sequence.

But also you allow the true B-primer to anneal and amplify on the other end of the amplicon during emPCR.

--
Phillip

[zhengz]

My guess is that using inosines to prevent supressive PCR.

Inosines allow forming of Y adaptors initially, but after the sencond cycle of amplification, it's sequence on the newly formed strand would be changed, such that it prevents formation of a haipin struction.

Incidently, I once designed a Y adaptor with a stem of lenght 17 bp (of course without inosines), which completely inhibit PCR amplification. From the sequences you provided above, I see that there is an 18bp stem. If inosines were not used, amplification would not work.

Later, I designed a set of new Y adaptors with stems of only 10 bp, which allows formation of Y adaptors initially and at the same time without inhibiting amplification. The 10 bp actually serve as MID sequences. It truned out good (700M bp in total in one Titanium run, evenly [+/- 5%] distributed for 8 MIDs).

[pmiguel]

My guess is that using inosines to prevent supressive PCR.

Inosines allow forming of Y adaptors initially, but after the sencond cycle of amplification, it's sequence on the newly formed strand would be changed, such that it prevents formation of a haipin struction.

Incidently, I once designed an Y adaptor with a loop of lenght 17 bp, which completely inhibit PCR amplification. From the sequence you provided above, I see that there is an 18bp loop. If inosines were not used, amplification would not work.

Actually, later I designed a Y adaptor with a loop of only 10 bp, which allows formation of Y adaptor initially and at the same time without inhibiting amplification. It truned out good (700M bp in total in one Titanium run).

I think you mean "stem", not loop. The stem is where the bases of a single strand of DNA or RNA anneal. The loop is the non-base pairing internal part of the strand.

Seems like as long as you can crank your concentration of PCR primers high enough, you can overcome the unimolecular kinetics of stem formation. Especially since the loops on these will be >400 bases most of the time. Large loops will destabilize the stems.

--
Phillip

[zhengz]

Sorry for confusing...I use the two words quite often recently and mixed them from time to time.

Thanks Philip. I did not know that before.

[Zaag]

That sounds very plausible, thanks for your input!

[pmiguel]

Sorry for confusing...I use the two words quite often recently and mixed them from time to time.

Thanks Philip. I did not know that before.

Sure.

By the way, your Y-primer design was exactly how I presumed Roche would implement Y-primers. It makes perfect sense.

The deranged, non-Euclidian method they did use nearly gives me a stroke every time I try to grasp it.

Were you going to tell us more about this 700 MB run? I've heard of people breaking 600 MB on a Titanium run, but never 700 MB.

--
Phillip

[zhengz]

At the first look at the result that gave 711 million bases (1.7 M passFilter reads, median read length ~430 bases, mode read length ~500 bases), I doubted I had made something seriously wrong and the results were not trustable.

After trimming away the homemade adaptor sequence (I had a Taqman-MGB probe binding site on the B branch), 699 million bases left.

Results of mapping and De novo assembly looked good - >99.5% of the read mapped to the right bacterial genome, >99.5% of bases were assembled. Then I was not so worried as in the beginning.

Still, I have no idea why Roche uses several inosies instead of simply shorten the stem, which, like you said, makes perfect sense. Maybe there are other reseasons to use inosies than the issue of suppresive PCR . But, at lease it works for me without using inosies adaptors.