Tm prediction : more

Bill Pearson wrp at cyclops.micr.Virginia.EDU
Thu Feb 11 14:00:00 EST 1993

In article <C26sMA.MLA at> camdna at (Darren Natale) writes:
>In article <C25JG9.837 at>, Robert.H.Gross at (Bob Gross) writes...
>>In article <1993Feb8.190445.1 at>
>>vldnaseq at writes:
>>> Naturally, Tm is DNA concentration dependent
>>Shouldn't Tm be DNA concentration INdependent? It just represents the
>>energy needed to pull the two strands of DNA apart - that is
>>independent of how many molecules of DNA there are. Reassociation of
>>the two strands obviously is concentration dependent.
>Tm is defined as the temperature at which 50% of the DNA molecules are 
>single-stranded. The strand separation/strand annealing steps are both
>reversible processes. Therefore, the ability of a single stranded molecule
>to become annealed to its complement will depend on the likelihood of
>"finding" that complement--thus it is concentration dependant.

	I think that this argument is somewhat misleading. As
typically measured, the Tm is the temperature at which 50% of each DNA
molecule, on average, is single stranded.  In a traditional optical Tm
measurement with a uniform local base composition, one would expect
that none of the molecules would be completely single stranded at Tm.

	Sometimes "Tm" is measured on hydroxyapatite columns or
membranes in a way that a strand is not considered single-stranded
until it is completely denatured.  This is also called Td - half
dissociated rather than half melted.  I would argue that the Td does
not measure an "equilibrium" phenomena, because once the DNA is
melted, it will almost never reassociate during the time course of a
typical melting experiment. We know that the optimal temperature for
reassociation ranges from 20 to 30 oC below Tm, at Tm there is no
reassociation. (Even at Tm-25oC, a typical DNA sample - say E. Coli
with a complexity of 4 X 10^6 - at a concentration of 50 ug/ml single
stranded, would take 10 hr to become 50% renatured. The actual rate
would be much smaller, because near the Tm at most 1/2 of the DNA
would be single stranded and, as I mentioned before, the reassociation
rate would be 0. The case for PCR primers is more interesting.  They
might be able to reassociate significantly near Tm - if there were a
complementary sequence.)  This is in contrast to an optical Tm, where
there is an equilibrium - on a DNA fragment the bases are melting and
renucleating until the two strands are separated.

	In practice with modest complexity DNAs, Tm is concentration

Bill Pearson

More information about the Bio-soft mailing list