Questions for SHARE-Seq

[SukuraAddiction]

Hi Dr. Chen,
Thank you for sharing detailed information about the various single-cell sequencing technologies! I’ve been studying the SHARE-seq method and have a question regarding the design of the Mosaic End (ME) primer. In the paper, the ME primer is described as follows:

I noticed that this sequence is 15bp long, while the Mosaic End (ME) is generally 19bp. Could you please explain why a shorter sequence was chosen in this case? Additionally, I thought the use of 3ddC would block the oligonucleotide from extending further, but I didn't see this effect clearly represented in the technical roadmap you shared on GitHub. Also, considering the ME sequence provided, should the Tn5 schematic also be updated to reflect this?
Thank you for your time and consideration. Your insights would be greatly appreciated!
Best regards,
Elena

[dbrg77]

Hi Elena,

Thank you for those questions.

In terms of the 15 bp Blocked_ME_Comp, you are correct. I think many people, us included, use 19 bp when assembling our own Tn5. I did not notice there ME_Comp was only 15 bp when I created the page, so good catch! The full 19-bp ME is still present in the top strand, but maybe the 15-bp dsDNA is already enough for the transposase Tn5 to bind and perform tagmentation. I don't have a clear answer, and we may give it a test in the future.

I just realised that they have now put the full protocol on protocols.io here: https://www.protocols.io/view/share-seq-protocol-v2-2-81wgbx1oylpk/v2

Do you want to ask this question there to see if they have more insights? I will update the Tn5 schematic in the Github once we got confirmation.

For the modified oligos, I kind of ignore all modifications in all methods here to make things simpler. I listed them in the adapter sequence section, and omit them during the roadmap, which makes alignment easier. I assume people who need to know those details would read the paper and get the rationale behind the modifications from the paper.

You just reminded me that I forgot to added the modified Blocked_ME_Comp in the SHARE-seq page. I will add that soon.

I hope this is clear.

Thank you for your feedback and let me know if yo have further questions.

Regards,
Xi

[dbrg77]

Hi Elena,

My student has done some test, and we can confirm that the 15-bp ME is working as efficiently as the 19-bp one. Please see the library profiles on a gel below. We also sequenced them and the metrics look the same. Therefore, I think we can conclude that the 15-bp just works. At this point, I still don't know why the 15-bp ME was chosen in the first place, but maybe we can get some idea by checking the structure of Tn5 binding to ME in PDB. Anyway, I have now changed the schematic view of the Tn5 dimer in the SHARE-seq page.

Thank you very much for pointing out this for me and I'm closing the issue now. Feel free to re-open it if you have more related questions.

Regards,
Xi

[SukuraAddiction]

Hi Dr. Chen,
My colleague and I have replicated the experimental procedures from SHARE-Seq and confirmed that the 15 bp ME indeed works. We believe that the shorter 15 bp ME sequence with full phosphorothioate modification results in a lower annealing temperature compared to the regular 19 bp ME. This makes it easier for the round 1 barcode to displace the ME sequence, especially since there is no template-switching step involved during the split-and-pool process.

As for the use of ddC modification, our hypothesis is that it may serve to prevent the displaced ME from acting as a primer, which could otherwise lead to non-specific amplification products.

I truly appreciate your help on this matter — thank you again.

Regards,
Elena