Hi,
You can do this with the 'filter' action. You can read in the
previously generated projections and the corresponding trajectory, and
filter for the frames that match certain criteria. Say for example you
see a free energy minimum at PC1 2.0 and PC2 4.0. You could do
something like:
readdata dih.project.dat name Project
parm myparm.parm7
trajin mytraj.nc
filter Project:2 min 1.75 max 2.25 Project:3 min 3.75 max 4.25
trajout filtered.nc
Here I'm giving a width of 0.5 Ang in each dimension but you can
adjust to the actual shape of your minima. Hope this helps,
-Dan
On Wed, Aug 10, 2022 at 4:19 AM ning <ningdancer.163.com> wrote:
>
> Hi Daniel,
>
> Thanks so much for your offered commands. Yes it works, now I can plot the free energy landscape that considers the temperature into the data processing. However, an new problem has came, that how do I extract the corresponding frames which are thought to be in the low free energy states. Now I have the gnu_plot file, but I don't know the mutual correspondences between frames of trajectory and gnu_plot file data. I tried to export the projection data along the top 3 eigenvectors, with this fomat: #Frame Mode1 Mode2 ... Please help me. Any suggestions are welcome.
>
> Sincerely,
> Ning
>
>
>
>
>
>
>
> At 2022-07-22 22:32:39, "Daniel Roe" <daniel.r.roe.gmail.com> wrote:
> >Hi,
> >
> >You want to create a 2 dimensional histogram of the projection data
> >along whichever eigenvectors you choose, e.g. (assuming
> >dih.project.dat contains your projection data in columns, with column
> >1 being the frame #, column 2 being the projection along the 1st
> >eigenvector, etc):
> >
> >readdata dih.project.dat name Project
> >runanalysis hist Project:2 Project:3 name bins 100 out project.ev.1.2.gnu
> >
> >This will write the 2d histogram in gnuplot format. *If* you are
> >certain you have converged data (and that is a very important "if")
> >you can add the 'free <temperature>' keyword to calculate
> >population-based free energy values for each bin at the specified
> >temperature.
> >
> >If you haven't seen this tutorial already you may find it helpful:
> >https://amberhub.chpc.utah.edu/introduction-to-principal-component-analysis/
> >
> >Hope this helps,
> >
> >-Dan
> >
> >On Tue, Jul 19, 2022 at 11:24 PM ning via AMBER <amber.ambermd.org> wrote:
> >>
> >> Hi, Amber community
> >>
> >>
> >> I tried to make a free energy landscape (FEL) about protein conformation changes during accelerated MD simulation. However, when trying the tutorial guidance (https://ambermd.org/tutorials/advanced/tutorial22/section3.php), some strange errors appeared. I don't know what happen about the version upgrade from amber14 to amber20. Anyway, I decided to use another way, such as dihedral angle method, to get the FEL plotting. Here is my cpptraj input commands:
> >>
> >>
> >> Here is part1:
> >> parm s.prmtop
> >> trajin s_prod_amd1.nc
> >> multidihedral BB phi psi resrange 1-58 out PhiPsi_1-58.dat
> >> run
> >> matrix dihcovar dihedrals BB[*] out dihcovar.dat name DIH
> >> diagmatrix DIH vecs 6 out modes.dihcovar.dat name DIHMODES
> >> run
> >>
> >>
> >> Here is part2:
> >> parm s.prmtop
> >> trajin s_prod_amd1.nc
> >> readdata PhiPsi_1-58.dat name BB
> >> dataset mode torsion BB
> >> readdata modes.dihcovar.dat name MyModes
> >> projection evecs modes MyModes out dih.project.dat beg 1 end 6 dihedrals BB
> >> run
> >>
> >>
> >> When it finished, I got these files: PhiPsi_1-58.dat, modes.dihcovar.dat, dihcovar.dat, dih.project.dat. Now, how do I convert the dih.project.dat into a dataset for FEL plotting using cpptraj or other programs? If some errors appeared in my commands, please remind me, thanks!
> >>
> >>
> >> Sincerely,
> >> Ning
> >>
> >>
> >>
> >> _______________________________________________
> >> AMBER mailing list
> >> AMBER.ambermd.org
> >> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 12 2022 - 10:00:03 PDT
