Force Matching Resulting Plots

23 views
Skip to first unread message

HET PATEL

unread,
Jan 28, 2025, 6:49:55 PMJan 28
to votca
Hello Everyone,

This is my first forum post so sorry if I mess something up. I am performing bottom-up coarse graining on peptide in water and am currently at the step where I need to perform non-bonded force matching. After performing force matching I get reasonable potential plots for all beads except 2 and I don't really know why this is the case. An example output for the non-bonded force matching for the A-B (seems correct) and C-E bead (seems incorrect) is attached. What are some possible reasons that would cause the potential graph shown in the C-E bead? I've tried increasing frames per block and sampling longer simulation time (~500ns).

Note the AB bead showns a proper well when zoomed in. It has high repsulsion on low radius values, which is expected.

table_A_B_16.png
table_C_E_16.png

Best,
Het Patel


Christoph Junghans

unread,
Feb 5, 2025, 12:53:35 PMFeb 5
to vo...@googlegroups.com
Sorry, I am not a force-matching expert (anymore), so I will leave this question to somebody else to answer, but looking at http://dx.doi.org/10.1002/mats.201100011, you might want to double check your exclusion list.

Christoph 

table_A_B_16.png
table_C_E_16.png

Best,
Het Patel


--
Join us on Slack: https://join.slack.com/t/votca/signup
---
You received this message because you are subscribed to the Google Groups "votca" group.
To unsubscribe from this group and stop receiving emails from it, send an email to votca+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/votca/a0666dec-7e44-4150-b167-5411bab8bd35n%40googlegroups.com.


--
Christoph Junghans
Web: http://www.compphys.de

HET PATEL

unread,
Feb 5, 2025, 1:51:56 PMFeb 5
to votca
Hello Chrisotph,

Thank you for the response. Quick clarification this is for a polymer in water, not peptide. It is a cumstom polmyer I drew in Avagdro2 and parameterized using Amber Antechamber. Regarding exclusions I want to make sure I understand it properly.....

From my understanding the exclusions in topology file are such that the molecule experiences no internal interactions (intramolecular) and should only experience interactions between other molecules. Hence, the exclusion list in the topology file should look like a tree. If you have 5 atoms in your molecules, then it would be:

1 2 3 4 5
2 3 4 5
3 4 5
4 5

In my case I have 96 atoms in my polymer. Hence, in the topology the exclusion list should be a triangle tree for all 96 atoms excluding interactions with each other. Also I should have deleted all entries in the [bonds] [pairs] [angles] [dihedrals] sections.

1 2 3 .... 96
2 3 4 ....96
....
94 95 96
95 96

Is this correct? Or do I have the wrong understanding?

Best,
Het Patel

Akash Banerjee

unread,
Feb 6, 2025, 9:17:21 AMFeb 6
to vo...@googlegroups.com
Hi Het,

I remember having the same issue. If I remember correctly, there is another parameter ( beyond the ones you mentioned) that you can tune. The parameter which defines the starting point (distance) for the FM potential was helpful to get correct potentials with the good features ( very high potential value at low distances and a nice well after that). I had written a script to automate tuning these parameters to get good potentials. 

Best,
Akash 

HET PATEL

unread,
Feb 6, 2025, 12:51:26 PMFeb 6
to votca
Hello Akash,

I have tried adjusting the <min> value in force matching as well as the <step> and <outstep> values. Unfortunatly it doesn't seem to fix the issue. I have decided to re-run atomistic simulation with a greater concentration to increase interactions bewteen polymers for better sampling. Is there anything else I can try in the meantime? Also, I would love to take a look at the script you have created. Is this on your GitHub?

Best,
Het Patel

Akash Banerjee

unread,
Feb 6, 2025, 4:01:54 PMFeb 6
to vo...@googlegroups.com
Hi Het, 

I basically did a brute force parameter screening: start * grid * frames. You can check out this script in my github repo. You can check out the readme for our FM approach. All potentials came out good, but some of them needed to be dealt with separately because of the issues you mentioned. They are placed in FM-Anomalies. Ideally, all potentials need to be resolved together to resolve the PMF correctly (I think this has been discussed in a previous thread in this mailing list). We could not do it because of computational limitations. Hence, our process is not completely correct but you may get some ideas to fix your potentials. 


Best,
Akash



--
akash
Reply all
Reply to author
Forward
0 new messages