UNAFold

Hello,

I am a member of the Small Biosystems Lab (University of Barcelona), led by Prof. Felix Ritort. Over the years, we have used the mFold server to compare with the results obtained in single-molecule experiments in our lab. However, recently we noticed that the prediction by Mfold changed with respect to the past.
In particular, we found a large discrepancy between the melting temperatures predicted by the DINAMelt Two-State Melting Hybridization tool and the experimental results reported by Owczarzy, R. et al. (Biochemistry 43.12 (2004): 3537-3554) for a set of 92 DNA oligos. Is the server using a different reference for the NNBP thermodynamic parameters rather than the UO set (Santalucia et al.)? There has been any other change (or mistake by us in using the server) that could help us understand this discrepancy?
Thank you in advance for your help.

Best,

Paolo Rissone

I apologize for not seeing this message sooner!

The DNA parameters have not changed with UNAFold 4.0 (although the RNA parameters have).

If I recall correctly, the Owczarzy data set was our primary reference when UNAFold was originally developed. I dug up what I think are those melting profiles and submitted one as a test:

http://www.unafold.org/results2/twostate/240128/155927/

It seems to agree fairly well with the measured values of ΔH=-248.55 kcal/mol, ΔS = -679.57 cal/mol/K and Tm=92.6°C. I wonder if you can share an example where there isn't good agreement?