Trimmomatic: What is the difference between paired and unpaired output files in paired-end mode

Question

[dllopezr]

Hi everyone

I am working with paired-end data in Trimmomatic. This software asks for Forward and Reverse reads and outputs four files: Paired Forward, Unpaired Forward, Paired Reverse and Unpaired Reverse.

¿What is the difference between the Paired and Unpaired files? ¿Which of them I must use for the next analysis?

[RT]

can we not mention unpaired ? i have a reverse read file that needs a lot of trimming but forward file is usable. I don't understand how to go about it

Answer 1

[Carlo Yague]

After trimming with trimmomatic, some reads can be discared, and their mate become "unpaired". Usually, only a small fraction of the reads become unpaired, and they can be ignored. Alternatively, you can map them as single reads along with the remaining paired reads.

[dllopezr]

Thanks, Carlo. So, in the case when there is not reads discarded ¿There is no difference between paired and unpaired reads?

[Carlo Yague]

When there is no reads discarded, all the reads should remain paired and the unpaired files should be empty.

[livia.tolve]

Hi, which proportion of unpaired reads is considered good for each sample in a paired end sequencing experiment?

[Carlo Yague]

I can't give you a number because the proportion of unpaired reads also depends heavily on the stringency of trimming parameters applied. For instance in you chose to keep all the reads (no filter), then you should obtain 0% unpaired reads regardless of the quality of your sample. Conversely, if you apply extremely stringent filters (not recommended in most cases), you might end up with relatively high % of unpaired reads, even if your sample is perfectly fine. All things being equal though, the lower unpaired reads, the better.