probe size: please help

Martin Leach leach at
Thu May 12 08:14:19 EST 1994

In article <CpoDKG.C3L at>, mes at (Mark Siddall)

> Aplogies for what may appear to be a naive query.  I am a molecular
> neophyte and hope to soon remedy that.
> Here's my situation:  I am armed with sequences for ribosomal genes for
> two related parasites.  I desire a probe that will light-up parasite1 if
> it is there but not parasite2 and not the host.  Notably, the ssu rDNA
> gene, as a whole is highly conserved across taxa.  On the other hand I
> can find regions of hypervariability (across species but not within a
> species).
> Here's my question:  for a descent probe, how long should it be?  Will
> 20 bp do it?  How much sequence similarity with the non-target parasite2
> and host can I have without non-specific binding problems?  And is
> A-T : G-C composition going to be a problem for specificity?  Ideally
> I want to keep the prob(es) short as I may use them for priming
> PCR reactions at a later date.
> Any help/pointers/redirections are most welcome!
> Thanks in advance.
> Mark

Dear mark

if hybridisation of the probe/washing is performed in alkylammonium salts
(see maniatis) then a single base-pair difference can be discriminated by a
5C difference in wash temp.

eg. 20bp sticking washes at 60C
    19 bp sticking washes at 55C

this is due to the alkylammonium salt concentration.

if u cant find it in the new maniatis - i have several ref.s on this.

In fact we used 20-25mer oligos to discriminate point mutations on pcr
products from cjd individuals.....

If you are interested in this - email me and i will post/mail some refs.



.....          Martin Leach                Email:leach at 
   _|____      Dept. of Pharmacology       Phone: (617) 638-5323        
   / o  /      Boston Univ. School of Med. Fax:   (617) 638-4329         
 _/  |-/__==/  80 E. Concord St. (L603)
(BULLDOZER) \_ Boston MA 02118            "Not the old underpants on your
               USA                           head.....WIBBLE" -BLACKADDER  

More information about the Methods mailing list