To improve homologous recombination efficiency

[Hong Qin]

For those who use homologous recombination, especially those working 
on knock-out, I would like to know what is your opinion about the  
methods we can use to improve the efficiency so far. 

First, the longer the arms, the higher the efficiency.

Second, the shorter the targetted knock-out region, the the higher the 
efficiency. 

Third, when use positive and negative selection, say neo and tk, the 
closer they are, the higher the efficiency.  

I am currently experiencing a very low homologous recombination 
efficiency.  I am considering to use two tk cassettes in a new 
construct. I heard some rumors that tk does not contribute too much in 
some cases.  Any suggestions are very welcome. 

Hong Qin
U of Chicago