Inosine in degenerate primers

The Great American Gene Company geneco at
Wed Feb 12 13:16:15 EST 1997

In article <33011E2A.4AE3 at>,
	johnson at ("Dr. Douglas A. Johnson") wrote:

>In the archives there are many E-mails concerning the use of inosine as
>a substitute for several bases when synthesizing degenerate primers. We
>would like to use this approach too; however, there is little mention of
>any problems or successes with the different, available
>thermopolymerases. When it comes to the use of inosine-containing
>degenerate primers, are all polymerases created equal??
>Doug Johnson

 Dear Doug:

There are a couple of things to keep in mind when making the decision to chose degeneracies over inosine, or vice
versa.  They have different impacts and it is useful to know what those impacts are so as to chose the best option
for a particular circumstance.

First, when you use degeneracies, such as R or Y or N, for example, you should pay attention to the "degeneracy
index".  This tells you how many different sequences you are dealing with.  For instance, if you have a single N in a
sequence, the degeneracy index is 4 and you have 4 sequences.  If you have 5 N's, however, you have 4 to the 5th
power of degeneracies.  The degeneracy index is 1024, i.e. 1024 sequences.  You can see that it would be easy to
quickly "water down" any individual sequence in the collection.  

Inosines present a very different problem.  We tend to think of inosine as "pairing with any base" but, in reality, it
should be thought of as "not refusing to pair with any base".  The distinction is important, since A-T and G-C pair
with negative Gibb's free energies (delta-G).  The magnitude of the "delta-G" is a measure of how strong the pairing
is, G-C having three hydrogen bonds pairs more strongly than A-T, etc.  Inosine "pairs" with any base with a
slightly positive delta-G.  This means that it is actually helix-destabilizing, although the magnitude of the delta-G is
so small that the destabilization is very minor.  But...  this means that you want to avoid things like two or more
inosines in a row, if you can.  You also want to avoid sequences that contain, for example,    ---IAI---  or ---IAAI--- 
or --IATI---  or -- IGI--  and so on, since one or two bases flanked by inosines (unless the two bases are GG, GC, or
CC) are really too weak to stabilize that portion of the helix.  In those cases, it would be better to use degeneracies,
or a combination of inosine and a degeneracy.

I hope that helps.  


More information about the Methods mailing list