simtk.org

Hello!
I am trying to model a stem-loop in SARS-CoV2 (stem-loop 2) and am running into some difficulties. I modified the tutorial input file and have gotten the base pairs to sufficiently pair, but I am struggling to get the rest of the structure to match a previously published homology model of this stem-loop. I have attached the input file, the last1.pdb, as well as the homology model of the stem-loop that I am trying to match relatively closely. My main problem is that I don't know what forces and parameters to apply and I couldn't find any input files or guides using this program in my literature search. Any insight or resources would be greatly appreciated!

Cheers,
Nicole

Hi Nicole,

Great application! Not sure which version of MMB you are using. Assuming it is a recent release, you can follow exercise 1B to apply NtCs:

http://pe1.scilifelab.se/MMB-annex/MMB.3_4.tutorial.pdf

To improve the helical structure I very strongly recommend adding Jiri Cerny's Nucleotide Conformers (NtCs). These will get you the backbone structure. The baseInteraction commands only restrain the bases, they do nothing to the backbone. Assuming you know nothing about the backbone conformation, the safest guess is that it is in class AA00 simply because it is the most populated class in A-form helices. NtCs are defined across dinucleotides. So the first two would be applied like this:


NtC A 48 49 AA00 .5
NtC A 49 50 AA00 .5

and so on. Don't forget to get the complementary strand.

Do you care about the loop structure? It might not be defined. If it is defined, then you could try to figure out if it is a motif of some sort, or maybe you could find a solved structure in the PDB that you believe has this loop. Then you could either extract the NtCs using DNATCO, or just homology model it within MMB (there are tutorials for that as well).

Sam

Hi Sam,
That was really helpful thank you! I am struggling to figure out how to model the loop without a homology model. What would you suggest if the loop is not defined? This particular loop does have a homology model but I am trying to test if I can get close to the homology model without constraining the loop.
Is this a case for physics where you want it?

Thanks!
Nicole