Two state melting(hybridization) problem

1 reply [Last post]
swimming
Offline
Joined: 07/09/2014

I use two-state melting on the website: http://mfold.rna.albany.edu/?q=DINAMelt/Two-state-melting

my inputs are 19bps RNA sequences, sense and antisense, to calculate the energy.
Sometimes in the result the 1st base pair are not paired. For example:

CGCGCGCGGCCCUGCGGGU and ACCCGCAGGGCCGCGCGCG

They are fully reverse complementary sequences, but the result from the website consider the 1st bps C1-G19 as single strand overhang rather than a paired base.

Could you tell me why this happened? Is there some biology behind this or is this some bugs?

Thank you.
best,
Ming

zukerm
Offline
Joined: 11/12/2010
Thank you for asking this

Thank you for asking this question. What you observed is not a bug. It is a result of the nearest neighbor free energy parameters. I don't believe there is any biology behind this. You should think in terms of statistical thermodynamics.

The terminal mismatch, C1∙G19, contributes to the stability of the helix through base stacking. It turns out that the free energy contribution of the mismatch is the same as if the helix ended with a base pair. In this case, with or without the terminal base pair are equivalent. You should think of this terminal pair as a mismatch 50% of the time and a base pair 50% or the time. You can upload the .ct file to the Structure-display-and-free-energy-determination link. This application plots the secondary structure/hybridization and annotates it with base pair probabilities. The probability annotation colors are shown here.