Which ForceField parameters?

[Robert DeLisle]

Perhaps a naive question, but I'm looking though the Bio3d documentation and I don't find a description of what forcefield parameters are used. What does Bio3d use for nma and aanma?

[Robert DeLisle]

Digging a bit further into the documentation, and I see that standard nma (outmodes = "calpha") using the calpha forcefield of Hinson. I don't see an indication of what is used for aanma when outmodes='noh', however. Is it still calpha? Should I specify something else when using all atom NMA?

[Yazhini arangas]

Even I am wondering. What are the cutoff and force field used and which method Bio3D is following etc for aanma. I have come across DIMB method (Diagonalization in a Mixed Basis) and it is given as a reference in aanma (rtb=TRUE) function. However I am not sure for aanma or aanma.pdb(s)() without rtb=TRUE option whether they use DIMB or only Hinsen method?. I feel for each option there should be proper reference since all references are put together (in help manual) it is difficult to figure out which reference is exactly used for what option in the function.

[Xinqiu Yao]

'aanma' is a new method and we have a brief introduction in our recent paper (https://pubs.acs.org/doi/10.1021/acs.jpcb.6b01991).

outmodes='noh' will output modes for all heavy atoms while outmodes='calpha' will output "effective" modes for C-alpha only. Both methods take an all-atom elastic network as input.

As I mentioned in our Email communications, Bio3D does not use DIMB for diagonalization (although we cite Tama et al Proteins 2000 but mainly for the rotation-translation block (RTB) method not DIMB that they implemented in the paper!).

[Robert DeLisle]

Thank you for the response, but I still can't determine what forcefield parameters are used for aanma.

[Lars Skjærven]

Yazhss: No cutoff is used. The pair force constants fall with r^-6. We used the same parametrization technique as in Hinsen et al 2000 and his c-alpha forcefield (cited in out paper). This is decribed in detail in the paper Xinqiu mentioned. The rtb approach is built on top of that.

[Lars Skjærven]

Kirk: see eq 1+2 in the mentioned paper.

see also help(load.enmff)