Question regarding 3' bias trend among Illumina, ONT, and PacBio (RNA-seq)

[jjbox]

I am interested in the 3' bias trend among Illumina, ONT, and PacBio, so RNA-seq in general.

When comparing 3' bias among three techs, my guess is ONT would have the most 3' bias, PacBio has the least, and Illumina should be somewhere in between those two techs. The reason I think this way is because long-read sequencing begins from the 3' end, and especially PacBio has a longer sequencing length (2200 ish bp) compared to ONT (660 ish bp). Thus, it should have less 3' bias compared to ONT. Illumina (90 bp) should be similar to or slightly worse than PacBio because obviously, it's shorter.

I want to understand the trend of 3' bias among these three techs and wonder what you guys think about my reasoning. Do you think my reasoning would be reasonable in general?

[GenomicSeq]

Hello jjbox I think your general idea of 3' bias makes sense for these technologies, but the sequencing lengths of PacBio and ONT have gotten much much longer recently. I think PacBio's HiFi sequencing can do 10-25 kb and ONT can do 10-100 kb depending on the run and application.

Also, I would think the 3' bias is very dependent on the library prep and fragmentation methods. There are a lot of fragmentation methods and amplification processes that cause a lot of bias in sequencing runs.