How ViennaRNA Package deal with character other than AUGC in RNA secondary structure prediction?

Question

[wyj.jackson]

I have a RNA sequence as following, where "o" connects two RNA sequences.

UAGGGoooooooAGUU

Most of the RNA secondary structure prediction tools do not allow characters other than AUGC, but I find out the ViennaRNA allows "o" and can predict structure with it. But I am wondering how ViennaRNA (RNA2Dfold) deal with "o". I can not find any clues on ViennaRNA website. Please help me if you know the answer:) Thanks in advance.

Answer 1

[Martombo]

non AUGC bases are considered part of the RNA molecule, but they cannot form any pair. they can be in a loop separating two matching stretches, like in the following example:

AAAAAAAAAAooooooooTTTTTTTTTT
((((((((((........))))))))))

which is not the same as if they were completely ignored:

AAAAAAAAAATTTTTTTTTT
((((((((....))))))))

can I ask you what do these o represent in your sequence?

[wyj.jackson]

Thanks for your reply.

Do you know how those non AUGC bases influence the energy model of hairpin, interior and multi loops in the package? Also treat them as unpaired bases?

For "o"s, our collaborator said that they are physically very flexible and can link two RNA sequences.

[Martombo]

according to how your collaborators described it, it looks like the ViennaRNA package deals with them as you want it to. They are in all respects like unpaired bases.