small DNA sample library prep

[vlee2]

Could somebody give me an advise how to prepare Titanium library using small DNA sample size < 0.5ug.
Thanks!

[gaster]

I'm not sure you will be able to do that with quantities that small. I would suggest you use genomiphi or something to amplify your DNA.


If you want to give it try, you should find some other way of nebulizing the DNA (as the yield is fairly poor using compressed Nitrogen). Try some kind of mechanical shredder. Downstream from that, your main problem will be seeing your DNA on a chip, but as long as you trust that you have the size range you want, then you can do the rest of the prep blind. You might have to use qPCR to accurately quantify it though.

See these references:

http://nar.oxfordjournals.org/cgi/content/full/36/1/e5

http://www.ncbi.nlm.nih.gov/pubmed/18550420

Good Luck!

[vlee2]

Thanks for your input gaster!
I also thought about using Covaris to nebulize DNA since it should give better fragments size distribution vs nitrogen nebulization. When I'll optimize the Covaris fragmentation, I can live without DNA chip.
During last step of ssDNA prep after melting away ssDNA I'm planning to use heat to disrupt streptavidin-biotin bond and thus increase the yield of ssDNA library.
What do you think about this: since the speed of reaction directly determined by concentration of reagents, and I use about 10 times less DNA then suggested by protocol, should I also increase the time of all steps during library prep?

[gaster]

I don't quite understand. I haven't done a Titanium prep yet, but previously you just wanted the non-biotinylated ssDNA. The biotin-bound strand was discarded. This step makes sure the adaptors are correctly bound.

If you do try heat to break that bond, I would definitely try to optimize it. I have found in the past that to break that bond you have to go fairly high (98 degrees).

A 454 tech told me that the real hurtle (as far starting quantity goes) is the correct sizing of the fragments. If you can be sure that you have correctly sized fragments, you can do the rest with really low amounts of template.

Good Luck.