What is the meaning of pair ends for de novo assembly?

[feralBiologist]

Hi,

Apologies for the newbie question but I find the concept of using paired end reads for de novo assembly somewhat confusing. From what I understand, the paired end reads are produced using a reference genome - is that correct?

But how can you pretend that you are doing a de novo assembly (e.g. agnostic of a reference) while utilising the paired end information? For me there seems to be a contradiction.

Thank you in advance for any responses!

--FB.

[kopi-o]

No, the paired end reads are not produced using a reference genome. They come from sequencing from both ends of a DNA fragment.

[boetsie]

Hi,

you might have a look at the attached picture and at an overview of sequencing at this website;

http://www.cbcb.umd.edu/research/assembly_primer.shtml

especially have a look at figure 2 there.

In summary;
DNA is copied and sheared into random fragments. Next, the fragments are size selected depending on the desired library. E.g. for a paired-end with 200bp insert size, fragments of approximately 200bp are kept (hence; there is always a deviation, so fragments of say 100 till 300bp are remaining). For a 500bp paired-end, fragments of 500bp are kept. From the remaining fragments both sides are sequenced, corresponding to the paired-end reads.

So during paired-end sequencing no reference genome is involved.

Hope this is clear,
Boetsie