[gmx-users] mixing of PCs for generating reduced free energy profile using g_sham

[R.S.K.Vijayan]

Hi Gromacs user community

a)  I was wondering if it is possible to mix  a common pair of PCs  from
two simulations (say, PC1 from a bound simulation vs PC1 from unbound
simulation or for that matter wild vs mutant)  as an reaction coordinates
to generate a comparative reduced free energy profile using the g_sham
utility.

b) Will such a approach  generate a meaningful comparative free energy
plot, akin to an conformational landscape plot commonly generated by  plotting
PC1 of one system vs. PC1 another system, to highlight the difference /overlap
in the conformational  sub-space sampled by the two systems

c) Given my understanding of underlying formula ∆G(R) =  -*KBT  *[ln P(R) –
ln P(max)]  as
implemented in g_sham i presume that it is impossible, since  mixing of PCs
from two different systems will not ensure a proper  probability
distribution, the data points has to be similar.



Thanking in advance


Regards,
Vijayan.R