Why the refined structure looks so messy?

[Leon Wang]

Hi, Bachir,

First, thank you for your comments and suggestion in previous questions. Here, I have some new questions.

Owing to the limits for this group, all figures are shared through Google Drive.

1. Why the refined structure look so messy close to the boarder of box? How can I improve it? Did I over-fit it?

Fig 1 and Fig 2 are the refined structures at the beginning and end, respectively. Attached are the related "pdb" files. 

Fig 1 Refined structure at the very beginning

https://drive.google.com/open?id=19vleOnnrAW4kbjcnY_-xgCQQMBfJ5UKA

Fig 2 Refined structure at the end

https://drive.google.com/open?id=17_w3Pr-vRfQiD1fjtQZnvaO_dyRxkpG5

2. I cannot output the RDF curves after certain iterations. I always get the error message shown below. Is there anything wrong in my fitting? I am still not confident that this fitting is good enough. Any comments would be highly appreciated.

  File "plotting-file-export.py", line 44, in <module>
    gcalc = PDF_CONSTRAINT.get_constraint_value()['pdf']   # computed G-array
KeyError: 'pdf'

Fig 3 displays the RDF at std err= 7587. After that  I am unable to output the RDF figure. 

Fig 3 RDFs at std err = 5787

https://drive.google.com/open?id=1CfP2Ak0To36zDTytjjHXM1fF0xd5InI-

3. There are some very sharp peaks existing in Fig 3. How can I restrain these peaks? I tested larger and smaller cutoffs, but they do not work well.

Any further comments would be highly appreciated.

Appreciatively,

Leon

[Bachir Aoun]

Hi Leon,

I can't really tell by the plots and figures why only the outer layers of your atomic structures seem messy. This can happen for many reasons here's a few

• you only created groups for the outer layer atoms
• your selector is not a RandomSelector but a weighted selector for instance
• your system is dense or your boundary conditions are off and only your outer layers group moves seems to enhance the engine total standard error

With this being said, it doesn't seem to me like a bug in fullrmc. I recommend you re-visiting your atomic system or experimental data

as for exporting the constraint. make sure you update to the latest version of fullrmc and use the constraints export helper method

PDF_CONSTRAINT.export(fileName=..., frame= ...)

Check the documentation for more information. what you did doesn't guaranty the constraint is computed. 

let me know if that works for you.

thanks

 

[Leon Wang]

Hi, Bachir,

Thanks for your instant reply. Yes. I will try it following your suggestion.

Best,

Leon

[Leon Wang]

Hi, Bachir,

My material is a nano material. In your previous reply, you suggest me check the boundary condition. The boundary condition is on. Do I need to set it off because it is nano material?

According to your suggestion and the parameter instructions of PairDistributionConstrain, do you think it is reasonable to add shapeFuncParams and windowFunction?

To my current knowledge, I did not created outer layer atom. Attached "run.py" file is shared from Peter.

Following your suggestion, I added PDF_CONSTRAIN.exprot(...) into "run.wly.py" which was run in the environment of Python2 and fullRMC3.3.I got error message like below:

Traceback (most recent call last):
  File "./run.wly.py", line 107, in <module>
    PDF_CONSTRAINT.export("constrain_track.log",frame=None, format='%s', delimiter='\t', comments='#')

TypeError: export() got an unexpected keyword argument 'frame'

Any suggestion would be highly appreciated.

[Bachir Aoun]

Leon,

There is no correlation between the shape function and the window function. It's reasonable to use both at the same time.

If you didn't subtract any solvent scattering intensity from your experimental data then you should set your boundary conditions. In the opposite, if you did subtract your solvent from your scattering data they your system can be considered isolated and therefore letting a shape function take care of your infinite boundary conditions.

starting from fullrmc 4 i have changed the way fullrmc plots and exports the constraints to allow selecting the frame.

you are using fullrmc 4.x syntax while using fullrmc 3.3

It's better if you stop using older versions of fullrmc and use my new releases

regards