gmx_MMPBSA_ana error

[suhrid saha]

Hey Everybody,

I recently install gmx_MMPBSA tools along with ambertools20. I ran the program on my protein-ligand trajectory which was generated by gromacs 2021. I calculated the PB calculation over 200 frames of my trajectory. After the calculation, I ran the gmx_MMPBSA_ana program by commanding,

gmx_MMPBSA_ana -f _GMXMMPBSA_info 

A windows pop up and after selecting "Accept" the window terminated by giving the error:

 

 OpenType support missing for "Ubuntu", script 12

Traceback (most recent call last):

  File "/home/pranta/.local/lib/python3.8/site-packages/GMXMMPBSA/analyzer/dialogs.py", line 266, in alljobs_finished

    self.parent.process_data(self.result_queue, self.options)

  File "/home/pranta/.local/lib/python3.8/site-packages/GMXMMPBSA/analyzer/gui.py", line 607, in process_data

    item.get_data()

  File "/home/pranta/.local/lib/python3.8/site-packages/GMXMMPBSA/analyzer/customitem.py", line 112, in get_data

    self.gmxMMPBSA_data = self.getplotdata()

  File "/home/pranta/.local/lib/python3.8/site-packages/GMXMMPBSA/analyzer/customitem.py", line 132, in getplotdata

    return_data.line_plot_dat = pd.DataFrame(data={'frames': self.frames[start:end:interval],

  File "/home/pranta/.local/lib/python3.8/site-packages/pandas/core/frame.py", line 529, in __init__

    mgr = init_dict(data, index, columns, dtype=dtype)

  File "/home/pranta/.local/lib/python3.8/site-packages/pandas/core/internals/construction.py", line 287, in init_dict

    return arrays_to_mgr(arrays, data_names, index, columns, dtype=dtype)

  File "/home/pranta/.local/lib/python3.8/site-packages/pandas/core/internals/construction.py", line 80, in arrays_to_mgr

    index = extract_index(arrays)

  File "/home/pranta/.local/lib/python3.8/site-packages/pandas/core/internals/construction.py", line 402, in extract_index

    raise ValueError("arrays must all be same length")

ValueError: arrays must all be same length

Aborted (core dumped)

I use the ubuntu 20.04.2 LTS operating system. 

Can anyone plz help me by saying what should I need to do.

With Thanks,

Suhrid Saha

[Mario Sergio Valdes]

Hi and welcome!

Please can you check if the variable startframe = 0? This is the most frequent mistake

Best!

Mario S.

[suhrid saha]

Thank you very much for the reply!!

yes, the startframe=0, endframe=7999, interval=40, verbose=2

numframes=200

is there any miscalculation? 

if so, can you please tell me What can I do next?

Regards,

Suhrid Saha

[Mario Sergio Valdes]

We already identified that bug and fixed it, but we haven't released v1.4.2 yet. Please see this issue to solve the problem (https://github.com/Valdes-Tresanco-MS/gmx_MMPBSA/issues/33#issuecomment-819075710)

The starframe variable always has to be >= 1

Best!

Mario S.

[suhrid saha]

Thank you very much for the solution!! 

it works.

But I have a new issue. In the Energy component (Avg.) figure, in the Y-axis the negative values are all upward and the positive values are downward. I am not getting the thing. (image1.png)

And in the Energy Component (delta Total) graph, the mol Avg(red line) started from 2000 frames instead of staring at the beginning. (image2.png).

Can you kindly help regarding this?

With Thanks,

Suhrid Saha

[Mario Sergio Valdes]

Everything in the graphics is correct. By default, we have put the negative values ​​upwards because thermodynamically it would be favorable. This is useful for the visualization per-residue, DG binding, etc ... We are currently working so that the user can change it according to their needs. In this case, their interaction is unfavorable, that is, a positive DH. Make sure that it is not due to factors not inherent to the system itself: that the trajectory has not PBC, the parameters of your simulation are correct, the structure defined in the -cs option is consistent, etc. You can review this section in the documentation (https://valdes-tresanco-ms.github.io/gmx_MMPBSA/Q%26A/calculations/)

In the 2nd case, it is also correct. The red line is the moving average (mv) with a window of 50 units. In your case, it processes a range of 8000 frames with an interval = 40, which is equivalent to 200 total frames. As the mv window is 50, it means that the first value corresponds to unit 50, that is, the first 49 frames do not have mv data. Since the interval is 40, 50 * 40 = 2000, as shown in the graph.

HTH!

Mario S.

[suhrid saha]

Thank you again to clarify my all doubts.

Best wishes to this tool and it is amazing. 

[Mario Sergio Valdes]

With pleasure and thanks to you for considering using it. 

If you have any questions, don't hesitate to tell us, we will try to help you as much as we can.

Best wishes!

Mario S.

[marioe911116]

Hi there! As I see you are computing your energy with PB model... You may want to try a different way to estimate the total non-polar solvation free energy. In this case, inp=2 is used which means the total non-polar solvation free energy is modeled as two terms: the cavity term and the dispersion term. The dispersion term is computed with a surface-based integration method closely related to the PCM solvent for quantum chemical programs. Under this framework, the cavity term is still computed as a term linearly proportional to the molecular solvent-accessible-surface area (SASA) or the molecular volume enclosed by SASA. You may want to try inp=1 and avoid the EDISPER contribution, where the total non-polar solvation free energy is modeled as a single term linearly proportional to the solvent-accessible surface area, as in the PARSE parameter set, that is, if INP = 1, USE_SAV must be equal to 0.

hope this helps!

[suhrid saha]

Thank you very much @marioe911116 for the suggestion.

But I have a question regarding the process you mentioned. If I approach INP=1 and avoid the EDISPER contribution, will this affect predicting the accurate binding free energy, as I am neglecting one term from the calculation? 

With thanks,

Suhrid Saha 

[Mario Ernesto]

As it is right now, your binding energy is unrealistic (positive value)... you can check this publication (https://pubs.acs.org/doi/full/10.1021/jp073399n) and see the drawbacks of modeling the total non-polar solvation free energy with two terms: the cavity term and the dispersion term. Sometimes there are imbalances in the cancellation of error between the two components and this can produce unrealistic non-polar energy values. Also, there are plenty of publications of MMPBSA with the non-polar term computed with only the cavity term. Have a go  and see how it works for your system...

best!

Mario E

--
You received this message because you are subscribed to a topic in the Google Groups "gmx_MMPBSA" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/gmx_mmpbsa/T8NOkmQgq50/unsubscribe.
To unsubscribe from this group and all its topics, send an email to gmx_mmpbsa+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gmx_mmpbsa/ae8d24fc-2118-4820-be64-b72e9cfc3655n%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[suhrid saha]

That's great! 

I will definitely try this out.

The help from your end is absolutely appreciable.

Thanks again