Does 1D complement-only read exist?

[ymc]

When I use poreminion fragstats to classify fast5's, theoretically I could use the 4th and 5th field to classify reads into 1D template-only, 1D complement-only, normal 2D and full 2D. In reality, after working with five datasets I downloaded from ENA, none of them contains any 1D complement-only reads.

I also wrote a simple python script to count template events and complement events but I never saw any reads that have zero template events but a positive number of complement events.

Does that mean 1D complement-only read doesn't exist???

[Ola]

They should not with the current chemistry since the motor protein sits on the adapter for the template strand.

[ymc]

They should not with the current chemistry since the motor protein sits on the adapter for the template strand.

Thank you very much for your reply.

So what is the difference between 2D reads and 1D reads that have both template and complement events? Could it be the former both template and complement come from the same molecule but the latter different molecules???

[Ola]

Thank you very much for your reply.

So what is the difference between 2D reads and 1D reads that have both template and complement events? Could it be the former both template and complement come from the same molecule but the latter different molecules???

The 2D read is basecalled from the signal from both strands so it is from the same molecule. Reads with failed 2D basecalls should be discarded as the hairpin may not have been correctly identified. For example, it could be that the template read includes part of the true complement. They will all be from the same molecule. Chimeric molecules are possible from the library prep though.

[ymc]

The 2D read is basecalled from the signal from both strands so it is from the same molecule. Reads with failed 2D basecalls should be discarded as the hairpin may not have been correctly identified. For example, it could be that the template read includes part of the true complement. They will all be from the same molecule. Chimeric molecules are possible from the library prep though.

Thank you very much for your reply.

Thanks for telling me that due to the unidentifiable hairpin adapter such that the boundary between template strand and complement strand is not clear, template events might also contain complement events. I suppose some template events might also went to complement events, correct?

What about the 1D template only reads? I suppose their hairpin adapter also couldn't be identified. But what else is different from 1D reads with template and complement?

Now that I think about this whole situation, I think a theory that reads are classified by how well hairpin adapter is identified might explain the whole thing quite well. 2D reads are reads with hairpin adapter location identified with high confidence. 1D reads with both template and complement events are reads with hairpin adapter locaiton identified with low confidence. And finally, 1D template-only reads are reads that hairpin adapter location couldn't be idenitified at all. So 1D template-only reads also suffers from the containing complement events problem.

Does this theory make sense?