I am trying to see the effect of side-chain fluctuation on molecular permeation through a transmembrane protein channel.
The idea is create a pdb file with a column of b-values related to the RMSF values of the MD trajectory.
I am using Lennart's script to do this, and everything is working fine.
What I tried was to set the b-values of all heavy atom main-chain atoms to 1.0 with the scalar command. Then, all the B side-chain values were set equal to the calculated RMSF value from the trajectory, using the correct formula relating RMSF fluctuation to b-value.
But, the above method blows up in visualization programs. I think this is probably due to the fact that all main-chain atoms are equal to 1, and the algorithm cannot handle this.
The putty representation takes the values from the b-factor of the CA atom. Theus, one should store ALL fluctuation values on the CA atoms, leaving all other atoms with 0.0, any other value or something like that. Larger RMSF fluctuations from side-chain residues would show up in the representation algorithm based on larger CA values.
I am using CHARMM to compute the RMSF of atoms over the trajectory, and the CHARMM script of Lennart computes the RMSF values ok, but it doesn't seem to be flexible enough to allow for adding the side chain RMSF to the CA RMSF. At least, my brain isn't flexible enough to do this task within the CHARMM scripting language. I have played around with the scalar wmain, wcomp, etc commands. But, wmain, for example, has separate locations for CAs and CBs, but how to add them together?
Can someone please provide me with a way to add the side-chain fluctuations to the CA atom fluctuations?
Thanks so much for your help.