My first few runs to determine microRNA structure

[kamlesh sahu]

Hello MMBians,

I am trying to get structure of a microRNA ( I found the connectivity using online server RNAstructure and used it in my input ). Here is the input -

---------------------------------------------
#sequence
RNA A 1 GUCAGAAUAAUGUCAAAGUGCUUACAGUGCAGGUAGUGAUAUGUGCAUCUACUGCAGUGAAGGCACUUGUAGCAUUAUGGUGAC

# stages
firstStage 1
lastStage 2

#general simulation parameters
baseInteractionScaleFactor 200
reportingInterval .5
numReportingIntervals 10
temperature 10


#base pairing forces to generate the hairpin
baseInteraction A 1 WatsonCrick A 84 WatsonCrick Cis
baseInteraction A 2 WatsonCrick A 83 WatsonCrick Cis
baseInteraction A 3 WatsonCrick A 82 WatsonCrick Cis
baseInteraction A 4 WatsonCrick A 81 WatsonCrick Cis
baseInteraction A 7 WatsonCrick A 78 WatsonCrick Cis
baseInteraction A 8 WatsonCrick A 77 WatsonCrick Cis
baseInteraction A 9 WatsonCrick A 76 WatsonCrick Cis
baseInteraction A 10 WatsonCrick A 75 WatsonCrick Cis
baseInteraction A 11 WatsonCrick A 74 WatsonCrick Cis
baseInteraction A 12 WatsonCrick A 73 WatsonCrick Cis
baseInteraction A 13 WatsonCrick A 72 WatsonCrick Cis
baseInteraction A 16 WatsonCrick A 68 WatsonCrick Cis
baseInteraction A 17 WatsonCrick A 67 WatsonCrick Cis
baseInteraction A 18 WatsonCrick A 66 WatsonCrick Cis
baseInteraction A 19 WatsonCrick A 65 WatsonCrick Cis
baseInteraction A 20 WatsonCrick A 64 WatsonCrick Cis
baseInteraction A 21 WatsonCrick A 63 WatsonCrick Cis
baseInteraction A 22 WatsonCrick A 62 WatsonCrick Cis
baseInteraction A 23 WatsonCrick A 61 WatsonCrick Cis
baseInteraction A 25 WatsonCrick A 59 WatsonCrick Cis
baseInteraction A 26 WatsonCrick A 58 WatsonCrick Cis
baseInteraction A 28 WatsonCrick A 56 WatsonCrick Cis
baseInteraction A 29 WatsonCrick A 55 WatsonCrick Cis
baseInteraction A 30 WatsonCrick A 54 WatsonCrick Cis
baseInteraction A 31 WatsonCrick A 53 WatsonCrick Cis
baseInteraction A 32 WatsonCrick A 52 WatsonCrick Cis
baseInteraction A 34 WatsonCrick A 51 WatsonCrick Cis
baseInteraction A 35 WatsonCrick A 50 WatsonCrick Cis
baseInteraction A 36 WatsonCrick A 49 WatsonCrick Cis
baseInteraction A 37 WatsonCrick A 47 WatsonCrick Cis
baseInteraction A 38 WatsonCrick A 46 WatsonCrick Cis

# Want to fold a GNRA tetraloop motif? Try the following:
# up above, change these parameters:
# baseInteractionScaleFactor 2000
# lastStage 2
# Then add these forces:
#readAtStage 2
#baseInteraction A 2662 Hoogsteen A 2659 SugarEdge Trans
#baseInteraction A 2660 Stacking3 A 2661 Stacking5
#baseInteraction A 2661 Stacking3 A 2662 Stacking5
#baseInteraction A 2658 Stacking3 A 2659 Stacking5
#readBlockEnd


numReportingIntervals 10
setDefaultMDParameters

----------------------------------------------

I am getting different structures and I don't know which structure to trust. I am attaching some PDBs. I played with numbers in input file and tried changing things like "reportingInterval 4.0" to "reportingInterval .5"

Or 'baseInteractionScaleFactor 200' to 'baseInteractionScaleFactor 10000' or 'baseInteractionScaleFactor 20000'
Similarly 'last stage 1 to lastage 2 or wastage 3

I have got following different types of folds (pictures attached). Alex very kindly confirmed on my email interactions that Ideally I should get structure similar to PDB 2N7X.

Could you please suggest what optimum parameter values should I use in input in order to get folds like the PDB file 2N7X.

Thank you
Regards,
kamlesh

[kamlesh sahu]

Sorry I forgot to attach pics of structures.
Here are the pics

[Samuel Flores]

Hi Kamlesh,

I am not sure I would recommend this approach. You can eventually get this to fold but it will be prone to kinetic traps. We got out of those using the Scrubber as described in the RNA and TCBB paper, but it took some patience. At the time it made sense because there was plenty of low-grade biochemical data and very little structural data. Now that there is more structural data I would suggest maybe doing homology modeling. 2N7X is as good a template as any, if you think this is what your structure looks like. There is a homology modeling example in the MMB Tutorial.

Sam

[kamlesh sahu]

Thank you so much Sam,

I will act as you suggested.

Thanks again,
Best Regards,
kamlesh