I tried this service to calculate temperatures. The nearest neighbour method consistently gives larger results for sequences like GCGCGC over sequences like CCCCC or GGGGG, while other methods don't give a difference. Is it nearest neighbour's feature or is there indeed a difference in melting temperatures?


The dependence of melting temperature on different dinucleotide pairs was experimentally measured; the meting temperature calculator is based on those results. Based on these results, you would see that a GC dinucleotide pair would differ from GG and CC pairs. Possibly, this is because of steric effects and stacking conformations.

SantaLucia (1998; author of the nearest neighbour model) has referred to different experimental studies from which the values were obtained, in the Table-1 of the paper.

  • $\begingroup$ That answers my question, but I am curious whether triplets and longer sequences of nucleotides have also been tested experimentally for melting temperature? $\endgroup$ – Sashko Lykhenko Mar 19 '17 at 2:14
  • 1
    $\begingroup$ @СашкоЛихенко I haven't checked if studies with triplets have been performed. However, I think that with longer stretches of nucleotides, there would be additional factors that would affect the binding (mostly arising from steric factors). For longer sequences (oligonucleotides) people do isothermal titration calorimetry (ITC) to calculate binding energy and other theromodynamic parameters. People also perform melting analysis by slowly increasing the temperature. This would provide the most accurate in-vitro estimate of melting temperature. $\endgroup$ – WYSIWYG Mar 20 '17 at 6:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.