Gromacs trajectory different when using mmgbsa

User discussions
1

Hello

Did anyone ever encounter an issue with mmpbsa regarding the trajectory frames being less than when using gmx check -f md_noPBC.xtc?
TrajError: start frame (1900) > total frames (1101)

2

You dont have frame 1900, you have only upto 1101, did you check the setting in mmpbsa script?.
what gives on gmx check?

3

Hi thank you for your response.

I have 3001 frames from my 30ns simulation in both md.xtc and md_noPBC.xtc when I checked using gmx check -f md_noPBC.xtc
I did not check the script but it’s gmx_MMPBSA that uses conda for the installation. It’s not the first mmpbsa.py but it’s a modified one.

The error message is
raise TrajError(‘start frame (%d) > total frames (%d)’ %
TrajError: start frame (1900) > total frames (1101)

4

i suspect the issue is with the mmpbsa script, since there is no issue in trajectory, recheck the selection and interval set in the script.

5

Thank you. I will do so.

6

Hi

Could you please help me understand what should be changed in the script for the frames to be correctly detected? It a yml file that then uses conda to be installed (Installation - gmx_MMPBSA Documentation)?

name:
channels:

  • conda-forge
  • defaults
    dependencies:
  • python=3.10
  • ambertools<=23.3
  • mpi4py<=3.1.5
  • gromacs<=2023.4
  • git
  • pip
  • pip:
    • pyqt5<=6.6.1
    • gmx-mmpbsa
    • pandas==1.2.2
    • seaborn<0.12
    • scipy>=1.6.1
    • matplotlib==3.5.2
    • tqdm
7

This is not where you need to change, you need to change the mmpbsa job script where you input your trajectory etc.
if i remmeber the file ends .inp or something.
It is better to write to the mmpbsa mailing list ( reccomended) or share your job script.

8

see here

image
image989×429 39.1 KB

these paramters you need to adjust

9

Hi

Thank you for your response. Yes I did adjust these parameters according to my number of frames. Let’s say I have 3000 frames and I want to start at 1900 and end at 3000. But it seems that it’s not able to detect the correct number of frames. From the log file, the number of frames are correct but while running it stops and says start frame is greater than total frames which is wrong.

10

Again, the issue is with mmpbsa and i believe it is in your mmpbsa script, please post it in their forum-along with the script, possibily you have a mistake in the script, time step or time or frame selection. or may you can cut the frames using gmx trjconc and use that for the input purpose-

11

Hi thank you for your response. Yes the error was in the script. I have been assisted by the author. It was related to the frame number used in nmode. Thanks.