Exercise 0 output

[Daniel S]

Hello,

I just started using the tool (thank you for it!) did a test run (Exercise 0), but my output structure seems to be wrong (please see attached). I did not change any default parameters and was running it on Mac OS and visualized it in PyMol. Are there additional parameters or files I should specify?

example_run.png (342.03 KiB) Viewed 2972 times

[Samuel Flores]

Hi Daniel,

Exercise 0 is just to make sure you installed everything and know how to invoke MMB and look at the output. It forms only one base pair and the steric clashes are prevented with some cheap collision detecting spheres, so it is not very physically realistic. It takes a little practice to look at it and recognize whether residue 1 is indeed base pairing with residue 6. Unfortunately you chose an orientation in pymol which makes it hard to tell how it went. You might try hiding all residues except 1 and 6. Most likely it is fine though.

What version of MMB are you running, and how did you install? I would recommend version 3.0 or 3.2, which are distributed as docker images. What is your objective with MMB if I may ask?



Sam

[Daniel S]

Hi Sam,

Oh I see, thank you for your reply! I did run the 3.0 version and it seems to have formed the pair (attached is the pic). Regarding the objective, I want to model some RNA sequence variants to see whether/how they might affect its structure. Some of those variants I would ideally want to model in complex with proteins. These are parts of ribosomal RNA together with ribosomal proteins, so I have reference 3D folds for them from published structures. I am thinking of using homology modeling or try to "create" a mutation and then model it?

Best,
Daniel

AU.png (204.13 KiB) Viewed 2946 times

[Samuel Flores]

Hi Daniel,

Absolutely! This is very doable. As I understand, you already have a complex of the RNA and protein together, but you want to make a mutation. So what I would do is prepare an input PDB file with your protein and RNA, you can call it e.g. last.1.pdb. Then in your commands.dat you would call:

firstStage 2
lastStage 2
loadSequencesFromPdb

You can use the substituteResidue command to mutate the protein. You can see an example of this in examples/commands.A_N.32.N.dat , this is the ZEMu example in the tutorial guide.

You can also mutate the RNA, but if you are going to mutate purine to pyrimidine or vice versa, you have to ignore one of the atoms, I think N1.. i think it's documented in the tutorial or ref guide.. we can look at it more once you get to that point.

Anyway this should be easy. whose lab are you in?

Sam

[Daniel S]

Awesome, thanks for the info. I think that's what I will try to do downstream once I explore the tool a bit more. I am a PhD candidate at NYU (https://hochwagenlab.bio.nyu.edu/people/). I have a decent dataset of rRNA mutations/variants so it will definitely be interesting to model them and possibly complement with in vivo data.

Daniel
ds5639@nyu.edu