PTMetaD-WTE post processing

[Kartheekpitta(PhD), IIIT-H.]

Hi all,
I  am using PTMetaD-WTE (exercise-4 https://plumed.github.io/doc-v2.3/user-doc/html/belfast-7.html) on a protein-DNA complex, in this sampling of two CV's has been monitored and four replicas are used at 500K,600K,700K and 800K. Four different HILLS files in REMD are obtained.  Visualization of individual trajectories obtained at 800K looks completely disordered in a 10ns run and  not able to observe any expected trend in the behaviour of CV. Hence how reliable are the individual free energy profiles to draw  any conclusion, if not how to construct a single free energy profile from the HILLS file obtained at four different temperatures. Any help will be highly appreciated.

[Max]

Hello,

well it is kind of expected that your system at 800K appears disordered. Typically PTMetaD-WTE (and other PT approaches) are used

in a temperature range from 300K (which is usually the temperature you are really interested in) to ~600K, with the high temperature

replicas used to cross free-energy barriers faster. 

While in principle the statistics accumulated at temperatures different than 300K can be reweighed to obtain information at 300K,

this procedure will not provide much information when the "hot" temperatures are so far away 

from the "cold" temperature (as in your case).

Max

[Massimiliano Bonomi]

And one more note.
If by disordered, you meant “discontinuous”, this is normal given the exchange
of conformations between neighboring temperatures typical of REMD.

Max

> --
> You received this message because you are subscribed to the Google Groups "PLUMED users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.
> To post to this group, send email to plumed...@googlegroups.com.
> Visit this group at https://groups.google.com/group/plumed-users.
> To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/82b38374-b95c-4699-8037-36fa35f20c5e%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Alejandro Gil Ley]

Hi

To be honest, I don't think you have a protein DNA TREMD with an appreciable, non negligible,  exchange acceptance when replicas are simulated at 500 and 600 K, how much is the exchange acceptance actually? 

Best

Alejandro 

On Mar 30, 2017 8:08 AM, "Kartheekpitta(PhD), IIIT-H." <karthe...@gmail.com> wrote:

Hi all,
I  am using PTMetaD-WTE (exercise-4 https://plumed.github.io/doc-v2.3/user-doc/html/belfast-7.html) on a protein-DNA complex, in this sampling of two CV's has been monitored and four replicas are used at 500K,600K,700K and 800K. Four different HILLS files in REMD are obtained.  Visualization of individual trajectories obtained at 800K looks completely disordered in a 10ns run and  not able to observe any expected trend in the behaviour of CV. Hence how reliable are the individual free energy profiles to draw  any conclusion, if not how to construct a single free energy profile from the HILLS file obtained at four different temperatures. Any help will be highly appreciated.

--

You received this message because you are subscribed to the Google Groups "PLUMED users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users+unsubscribe@googlegroups.com.

To post to this group, send email to plumed...@googlegroups.com.
Visit this group at https://groups.google.com/group/plumed-users.

To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/6a087cfc-6e22-4867-9903-964bc15b156d%40googlegroups.com.

[Massimiliano Bonomi]

Well, that would depend on the bias factor used on the Energy CV
Max

> On Mar 31, 2017, at 04:56, Alejandro Gil Ley <gil8...@gmail.com> wrote:
>
> Hi
> To be honest, I don't think you have a protein DNA TREMD with an appreciable, non negligible, exchange acceptance when replicas are simulated at 500 and 600 K, how much is the exchange acceptance actually?
> Best
> Alejandro
>
>
> On Mar 30, 2017 8:08 AM, "Kartheekpitta(PhD), IIIT-H." <karthe...@gmail.com> wrote:
> Hi all,
> I am using PTMetaD-WTE (exercise-4 https://plumed.github.io/doc-v2.3/user-doc/html/belfast-7.html) on a protein-DNA complex, in this sampling of two CV's has been monitored and four replicas are used at 500K,600K,700K and 800K. Four different HILLS files in REMD are obtained. Visualization of individual trajectories obtained at 800K looks completely disordered in a 10ns run and not able to observe any expected trend in the behaviour of CV. Hence how reliable are the individual free energy profiles to draw any conclusion, if not how to construct a single free energy profile from the HILLS file obtained at four different temperatures. Any help will be highly appreciated.
>
> --
> You received this message because you are subscribed to the Google Groups "PLUMED users" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.

> To post to this group, send email to plumed...@googlegroups.com.
> Visit this group at https://groups.google.com/group/plumed-users.
> To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/6a087cfc-6e22-4867-9903-964bc15b156d%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>
>

> --
> You received this message because you are subscribed to the Google Groups "PLUMED users" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to plumed-users...@googlegroups.com.

> To post to this group, send email to plumed...@googlegroups.com.
> Visit this group at https://groups.google.com/group/plumed-users.

> To view this discussion on the web visit https://groups.google.com/d/msgid/plumed-users/CAFh00goEQ%2Bym9GU3OhFmWdx1k3%3DfXs_67vf595_TnsavxjLZKw%40mail.gmail.com.

[CY19D501 Albin Joy]

Hello all

I have used a 12 replica, two-dimensional PTMetaD-WTE for a protein with ~2000 atoms (between 290K and 550K). I am interested in its unfolding studies and I have the following concerns regarding the analysis.

1) Whether I should demux the trajectory to use it with GROMACS analysis tools (like RMSD, SASA, Rg etc).

2) While using plumed sum_hills to calculate free energy, free energy even extends to negative values of CVs, ie, shows some free energy value at negative RMSD or negative Rg etc, which is physically impossible. (I wonder how RMSD and Rg can be negative.)

3) Also regarding T-REMD, I have observed greater exploration of RMSD, Rg etc when I restarted another crashed T-REMD with the last *.gro file output by GROMACS (the last gro file in each replica is actually reaching there by diffusing through various replicas, as the case with T-REMD). Is there any issue with extending  the T-REMD, with such a gro file and concatenating with the previous trajectory file?

Looking forward for help and thanking you,

Albin

[YAGYA CHAUDHARY]

Hi Albin,

Did you find the answer to your 1st and 2nd question? I am also having the same problem while running PTMETAD-WTE for a protein. If you know the solution to your questions, could you help me with this?

TIA

Yagya

[Massimiliano Bonomi]

Hi Albin,

please see answers below!

> 1) Whether I should demux the trajectory to use it with GROMACS analysis tools (like RMSD, SASA, Rg etc).

It is good practice in REMD simulations (and PT as well) to demux trajectories and calculate the time series of your CVs across the replica ladder.
This is done to make sure the simulation is converged and that the continuous walkers are not trapped in some regions of the CV space.

> 2) While using plumed sum_hills to calculate free energy, free energy even extends to negative values of CVs, ie, shows some free energy value at negative RMSD or negative Rg etc, which is physically impossible. (I wonder how RMSD and Rg can be negative.)

This might happen depending of the widths of your Metadynamics Gaussians. The value of the distance or RMSD CVs will never be negative, but if close to zero when you integrate the Gaussians with finite
width, you will have a non-zero bias potential (-> free-energy) also at negative values. If you search the mailing list, you will see multiple instances of this.

>
> 3) Also regarding T-REMD, I have observed greater exploration of RMSD, Rg etc when I restarted another crashed T-REMD with the last *.gro file output by GROMACS (the last gro file in each replica is actually reaching there by diffusing through various replicas, as the case with T-REMD). Is there any issue with extending the T-REMD, with such a gro file and concatenating with the previous trajectory file?

I usually restart using the state.cpt files, but in principle you can use the .gro files, making sure you are using the same simulation step across the different replicas.

Max