Making library from edge of sheared DNA smear - possible biases?

[Noa]

Hi
We used a Covaris machine to shear BAC DNA that is supposed to go on a 2x250bp MiSeq run. The specifications used were as recommended by Covaris website:

Target BP (peak) 500:
Peak incident power(W) - 175
Duty factor - 5%
Cycle per burst - 200
Treatment time(s) - 35

An inital run of two BACs gave the expected smear around the 500bp size. We then went ahead and prepp'ed 24 BACs and got smears more around the 250bp size.

We would like to know whether it is still OK to use this sheared DNA to make a library for an Illumina run, where the insert is size selected to be 500-700bp. Our main worry is that since we are size selecting the "edge" of the smear and not the peak, we are somehow biasing against certain DNA regions which will then harm our attempts to assemble the BAC in the downstream bioinformatic analyses.

Many thanks
Noa Sher

[mcnelson.phd]

Because the Covaris uses acoustics to mechanically shear the DNA, you technically shouldn't see any bias like you might see if you were doing an enzymatic or transposon (Nextera) type fragmentation.

By size selecting well about where most of your fragments would be, you could have issues with really poor yields though, to the point that you'd have to either do more rounds of PCR or have to just remake the libraries again.

Even on a MiSeq, there's no harm by sequencing small fragments where read 1 and read 2 completely overlap. In fact, this happens quite often with Nextera sample prep, so you just have to be aware of sequencing into the adapter on the other end of the fragment and making sure you trim that off before you try doing your assembly.