Do you size select??

[James]

Hello all,

I am going to be making some paired end libraries very soon.

I have already size selected my DNA prior to going through the library prep protocol, the DNA I have is ~ 50-500bp. So Size selecting before PCR would not remove any self-ligated adapters ( which I understand would be ~130bp ), and size selecting after PCR ... I actually don't understand the rationale behind a further size selection. But a PCR-purification column would remove primers from the PCR.

Thanks in advance for any input.

Cheers, J

[josdegraaf]

Main reasons for size selection are removing self-ligated adapters plus getting a tight as possible band for the paired end reads.

[RNAseqer]

I know some people are moving away size selection if they have a tight band when fragmenting their material. Anyone experience this?

[TUCF JSS]

Depending on the type of machine you are using to sequence (Hi-Seq, 454, etc..) different sized DNA inserts are preferred. I work mostly with Illumina sequencing on the GAII and Hi-Seq, and the ideal size insert is 200-400bp. Too much outside of that range and you will have poor Brige Amplification, and too small leads to redundant sequences.

If you are doing low throughput you may want to do a gel purification after the ligation step so that you can select a very tight band which is greater than the size of your adapter dimers (~130-150) but within the range of the preferred sequencing size (we usually select ~300BP). The PCR step therefore will only enrich this small range.

If you are doing high throughput, or do not wish to go through the time consuming process of a gel, AMPureXP (a bit expensive compared to gel/column purifications, BUT much more efficient) can help you with size selection. They use salt/PEG buffer as crowding reagents to preferentially bind DNA of certain sizes. Therefore, varying the amount of buffer present (IE the amount of bead solution used) will change what size DNA you will be purifying. By performing titration experiments on a 50BP ladder at a fixed volume (changing only the bead volume) you can run a quick gel and identify the amount of beads necessary to find your fragment size of interest.

I hope this helps! Best of luck.

~Jimmy

[protist]

[QUOTE=James;27755]Hello all,

I am going to be making some paired end libraries very soon.

I have already size selected my DNA prior to going through the library prep protocol, the DNA I have is ~ 50-500bp. So Size selecting before PCR would not remove any self-ligated adapters ( which I understand would be ~130bp ), and size selecting after PCR ... I actually don't understand the rationale behind a further size selection. But a PCR-purification column would remove primers from the PCR.

Even with the newest TruSeq gDNA kits Illumina still recommends gel based size selection. For the TruSeq RNAseq kits the size selection is bead based. Self-ligated adapters will suck up sequencing reads and the tighter your library band size the better the clustering. See the attached screen capture from the newest Illumina TruSeq gDNA protocol. Good luck with the libraries

[TonyBrooks]

We found that shearing on the Covaris with the following settings

Duty Cycle 10%
Intensity 5
Cycles per Burst 200
Time 6 cyles of 60 seconds each
Set Mode Frequency sweeping
Temperature 4°C

along with ALL clean ups using 1.8X volume Ampure XP beads gives us a tight enough range of fragment size. We try to avoid gels as they are expensive in both time and reagents if you need to process multiple libraries. We see no adapter dimers at the end.

[RNAseqer]

how wide of a size range are you able to get away with before sequencing? 300-400 (100 bp) or 300-500 (200 bp) or even more? how does the larger size cause problems with pe reads?