Help needed in screening libraries.

Steve Coon
Fri Jul 21 22:37:56 EST 1995

mbxrh at (Richard Hastings) wrote:
>Does anyone have any idea how many plaques I need to screen to have a
>good chance of finding a full length gene that I'm after?  It's a moth
>cDNA library cloned in lambda phage.  Also how many plaques is it best
>to use on a single plate of 150mm diameter?
>Thanks in advance for any help you can give me.

Well that is difficult to say. I personally think that only about 10,000 
appear on a plate of that size when they are as close together as 
possible and are just barely able to be seen. What part of the moth is 
the library made from? For example if you where trying to clone the 
insulin gene from a tissue it would be better if it were a tissue that 
secretes it and then it might even be one of every ten plaques! Certain 
receptors though native only to certain tissues might be one in 200,000 
genes with means only one plaque in 20 of those plates. Plus many of the 
so called "positives" that you will pick will turn out to be artifacts. 
Look for teardrop shaped colonies on your autorads. They tend to be the 
real thing. This is because as you tear of the filter paper from the 
plates the colonies tend to smear a bit. 

	It depends on the specificity of your probe. The longer and more 
specific your probe is the fewer positives you will pull out but the 
more likely they will be full lenght and the sequence you want. 

Use the Maniatis Molecular Cloning Manual or Current Protocols for 
protocols. They work well and include all the details. 

Are you using a amplified or unamplified library? An amplified library 
is one which the clones are allowed to multiply. The shorter clones 
begin to outnumber the longer ones and it may become harder to pull out 
a full lenght gene of great length.

Make sure that you store your library properly. It will degrade over 
time making it more difficult to clone full length transcripts and will 
obviously require more plates. (or make it impossible for you).

I hope some of this is helpfull for you. You can send me EMAIL if you 
have other questions. I don't read the newsgroup very often 

More information about the Methods mailing list