Hi everybody, I'm sequencing human samples in order to do some peak callings and genome coverage analysis. But I can't decide the best option for sequencing, hope you may help. I'm using NextSeq 500/550 High Output Kit v2 (75 cycles) so I have two options:
1) I can get 75 bases for SE but the samples have to be run in different flow cells (I don't like this option because of the error that it can generate
2)I can run all the samples on the one flow cell but as a paired end. However, this would only pick up 37 bases in each direction (74 bases in total) (less coverage)
For my analysis, I think that PE 37 bases should be enough, and also, I could extend the read length during the alignment.
Does any body can give me some information to justify this decision?
Thanks in advance,
The HO flow cell has 400mio clusters. It does not matter if you sequence each cluster in single or paired mode. Therefore, I do not understand why you think that in 2) you can pool your samples if using paired-end mode. Can you explain?
What kind of errors? If all your samples are in pool it can run on any number of FC.
If you are not looking for spatial information provided by paired-end reads then running the FC as 75 bp SE will give you longer reads that you can be confident about mapping to genome. I am not sure if you are referring to coverage as in knowing where the full fragment maps as opposed to how many reads cover a given base.
Is because of the index. It is not be able to load all samples on the one flowcell as they would only detect the first index and these are repeated. So I have to run the samples in two flowcells for single end (and this can introduce batch effect). For PE, as it can pick up index1 and index2 (are different) is possible to run them in the same flowcell. With coverage I mean the coverage for my fragment (37 bases PE or 75 bases PE) Do you think that going with SE is the best option? Thanks!
NextSeq 75 bp kit has enough reagents to do 2D index runs. You can run 75 bp X 8 bp X 8 bp (single end 2D run) and run them all in one FC.
I didn't prepare the library and the responsible for sequencing is telling me that for SE the samples have to be run in two different flowcells... so I think that these are the options that I have. regarding that, what is the best?
Without additional information I am not sure why you are being told that. If samples have 2D indexes then you should be able to run a single-end 2D run allowing you to get longer reads and all of your samples on the same FC. It is certainly possible with MiSeq/HiSeq and should be on NextSeq.
That said, running same pool on two FC should not generate unexpected "errors". If you are worried about batch effect then you can model that into your downstream analysis.
Lila M : Please do not delete posts, especially when they have comments/answers.
As the problem was solved I want to close the post!
closing
is an action moderators use to deal with incomplete/wayward/off-topic posts. Accepting an answer is enough action from you to provide closure to a post.