What is empirical evidence for mfold showing indel mismatch more destabilising than substitution ?

1 reply [Last post]
Joined: 01/23/2014

I used the DINAMelt Web Server to calculate melting temperature for DNA heteroduplexes. I use it for design of mismatched primers. The calculations indicate, that single nucleotide insertion or deletion lower the Tm of heteroduplex more strongly, i.e. are more destabilising, than substitution mismatches. What is the empirical evidence for this ? How did mfold come up with such calculations ?

I googled up this article, which seems to detect, that substitution and indels have *similar* effects on lowering heteroduplex Tm: Lipsky et al. DNA melting analysis for detection of single nucleotide polymorphisms. Clin Chem. 2001;47:635-44.

Other evidence probably exists, but seems googleproof. Can somebody give me the info, please ?

Joined: 11/12/2010
It is difficult to reply to

It is difficult to reply to your question because you give little information. Which application did you use on the DINAMelt web site? Did you compute full melting profiles? Are your mismatched primers very close to being perfectly complementary? I will assume that this is the case.

The nearest neighbor free energy parameters that we use were derived from melting studies on many different duplexes. For DNA, John SantaLucia published a series of articles (around 1998) where the effects of single mismatches were carefully measured. Based on these measurements and the resulting parameters, indels are (in general) much more likely to destabilize a duplex than single substitutions. I am not surprised at what you describe. What surprises me is what you report about substitutions and indels having similar effects on lowering Tm. I would have to read the article myself. There are very sensitive tests for detecting SNPs in DNA. For example, molecular beacons can work very well because there is a competition between hybridization and folding of the beacon, but that another topic altogether.