Mixed paired end and single end reads

20 views
Skip to first unread message

ullo...@googlemail.com

unread,
Oct 7, 2021, 9:27:17 AM10/7/21
to NGLess
Dear ngless-users,
I just received data from an experiment where two batches were sequenced with paired end sequencing and three single end runs. However, I'm curious how you would handle that. Reading in two file lists with load_fastq and then handling them together or preprocessing them separately and just run the downstream analysis together? Would you expect a mayor bias?

Cheers,
Ulrike

Luis Pedro Coelho

unread,
Oct 19, 2021, 5:09:03 PM10/19/21
to Ulrike Löber, NGLess List
Hi,

Sorry for the delay in replying (got lost in the inbox)

I would just put them all together in the same directory and load them as a single sample.

I would not expect that it would make much of a difference, tbh, exactly how this situation is handled. I would actually use this as a quality check: take a subset of the samples and process them as if the batches were different samples and see if it makes much of a difference.

HTH
Luis

Luis Pedro Coelho | Fudan University | http://luispedro.org
--
You received this message because you are subscribed to the Google Groups "NGLess" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ngless+un...@googlegroups.com.

Ulrike Löber

unread,
Oct 21, 2021, 1:29:21 PM10/21/21
to lu...@luispedro.org, NGLess List
Dear Luis,
The batches are indeed different samples, that's why I was worrying about the comparability.
Yours sincerely,
Ulrike


Am Di., Okt. 19, 2021 at 23:09 schrieb Luis Pedro Coelho

ullo...@googlemail.com

unread,
Nov 22, 2021, 7:07:36 AM11/22/21
to NGLess
Dear Luis,
how would you merge such kind of data? Creating two objects (preprocessed single and preprocessed paired end) and then merging them for the mapping process? How exactly can one write that up, since two sample lists must be used.

Yours sincerely,
Ulrike

Luis Pedro Coelho

unread,
Nov 25, 2021, 5:55:32 AM11/25/21
to Ulrike Löber, NGLess List
There is currently no simple way to do this in NGLess (although another person asked me about it this week, so it may not be that particular of a use case).

The only way is to merge them at the very beginning, so that you have a sample which includes all of the FQ files together. I guess the issue is that doing it naively would result in duplicated computation because you would be processing the same FQ file as part of the combined sample and as part of the individual one, which is very sub-optimal.

Another way is to merge the SAM files manually outside of NGLess and then load the combined samfile in ngless.

Best
Luis

Luis Pedro Coelho | Fudan University | http://luispedro.org


Reply all
Reply to author
Forward
0 new messages