how to ligate random oligos to vector to construct a library? - (Jan/17/2007 )

hi everyone, I am trying to ligate a pool of different oligos (50bp) with identical flanking ends to a double digested vector. Ologos are first annealled, klenow treated, and digested. Vectors (5.8 kb) are double digested, dephosphorylated. Ligation was performed o/n with T4 ligase, and electroporated into DH5-alpha.
However, I need to construct a large library, which can give me at least 1 million different colonies. Cause I have 18 bp random region on the oligos. But I only got 10,000 colonies from the transformants.
Who can give me some suggestion on how to increase the efficiency or new methods?
Thanks a lot!

How are you annealing these oligos? If you have random oligos, they will not anneal well, since it is statistically unlikely for a particular oligo to find its complementary partner (there may not even be one!). So, if you are mixing oligos, "annealing" and then trying to cut, I would be surprised if things cut.

A better (and even perhaps cheaper) approach would be to make random oligos with a constant 5' end, a random 18 bp middle, and a constant 20 bp 3' end. Then, make the complementary 20 bp oligo matching the 3' end, anneal and klenow extend. You now have dsDNA with a random internal region, but perfectly annealed. Cut this with restriction enzymes at sites in the 5' and 3' ends, and ligate into your vector. You'd want to avoid high temperature denaturing conditions after the klenow step, but it would be important to get rid of the klenow and dNTPs prior to the RE digests.

I did exactly the same thing. But didn't purify oligos after klenow, RE digests.
So you think Klenow and dNTP have negative effect on RE digests?

(table adapted from) Guide to Bacterial Transformation with Competent and Electrotransformable Cells, Life Technologies, 1995:

1.4e10 (Electroporation) 14,000,000 colonies/ng ligated DNA = 10-300 ng DNA in 20 µL (plateau transformation efficiency) *
1e10 Transformation efficiency (Electroporation) 10,000,000 colonies/ng ligated DNA = 10-300 in 20 µL*
5e9 (Ultracompetent) 5,000,000 colonies/ng ligated DNA = 10-300 in 20 µL
1e9 (Supercompetent) 1,000,000 colonies/ng ligated DNA = 10-25 in 100 µL**
5e8 500,000 colonies/ng ligated DNA
5e8 (Hanahan) 500,000 colonies/ng ligated DNA
1e8 (Hanahan) 100,000 colonies/ng ligated DNA = 10-25 in 100 µL
5e7 50,000 colonies/ng ligated DNA)
1e7 (RbCl2) 10,000 colonies/ng ligated DNA = 10-25
5e6 5,000 colonies/ng ligated DNA
1e6 (CaCl2) 1,000 colonies/ng ligated DNA = 20-25 in 50 µL
5e5 500 colonies/ng ligated DNA = 25
1e5 100 colonies/ng ligated DNA = 25
1e4 10 colonies/ng ligated DNA = 25 ng in 50 µL

* The reaction begins to saturate at 10-50 ng of ligated DNA in a 20 µL transformation. Reduced transformation efficiency may result when more DNA is added.
** The reaction begins to saturate at about 10 ng of ligated DNA in a 100 µL transformation.

Thanks!
this helps.

Yes, Klenow + dNTPs is fatal to restriction digested DNA. It will extend the RE overhang and blunt the end, making it impossible to ligate with a properly cut plasmid.

thanks very much!