General questions concerning Bias Exchange Metadynamics for Trp Cage protein folding

[casalini...@gmail.com]

Dear all,

I am performing BEMD of Trp Cage protein as a preliminary step, to check if I understood the basis of the approach.

I am using Gromacs 2018.3 + PLUMED 2.5.

Concerning the force field, I am using the a99SBdisp developed by Robustelli and coworkers, which is suitable for this protein according to the reference paper.

I am using 8 replicas biasing 7 CV:

1) Alpha carbon contacts;

2) Gamma carbon contacts;

3) Backbone H bonds;

4) Dihedral correlation;

5) Alpha helix content;

6) Parallel beta sheet content;

7) Antiparallel beta sheet content

Hills height is 0.1 kJ/mol, deposited every 500 steps (1ps). Random exchange attempts every 5000 steps (10 ps). 

I use as a reference paper the original manuscript from Piana and Laio, where they introduced BEMD with Trp Cage protein.

I have some general questions:

1) I have checked the files available in PLUMED NEST. Deighan and coworkers used PTWTE MetaD to investigate the folding of the same protein. Using the COORDINATION CV for Hbonds they adopted also the keyword "PAIR". Is there a specific reason?

2) The definition of DIHCORR CV on Plumed website and in the reference paper from Piana and Laio is different. Is there a typo on Plumed website or the implementation of this CV in PLUMED is different?

3) Concerning Alpha Helix/Parallel Beta Sheet/Antiparallel Beta Sheet content, the 1D FES show a minimum in 0, while it could be nice for beta sheet content but not consistent for alpha helix content (even after 300 ns per replica). In addition, these three replicas showed a very low exchange rate if compared with the other ones. Could it be a bad balance in the added bias amongst replicas?

I thank you all for you help and support.

With my best regards,

Tommaso

[palani sathish]

I am trying to calculate the Alpha carbon contacts. 

Please suggest to me, how to calculate the carbon alpha toms for two monomers?

Best regards,

Kandhan Palanisamy