Re: [COOT] [EXTERNAL] Re: Bug in coot to change torsion angles?

2022-08-30 Thread Gong, Zhen
Hi Paul,

I think it is a bug. I was able to rotation O10 along the C07-N08 bond, but not 
the O9. Here is what I got:
[cid:image001.png@01D8BCB0.38CC78E0]

Color-by-chain seems normal to me.

Zhen
From: Paul Emsley 
Date: Tuesday, August 30, 2022 at 20:25
To: Gong, Zhen , COOT@JISCMAIL.AC.UK 

Subject: Re: [EXTERNAL] Re: Bug in coot to change torsion angles?


On 31/08/2022 01:17, Gong, Zhen wrote:
Hi Paul,

Thank you very much for your timely reply. There are still some puzzles for me.

I would like to change some torsion angles of a ligand in coot but it did not 
work as what I expected. I was able to play with the torsion angles formed by 
C06 C07 N08 O10 but not C06 C07 N08 O09 (shown in the figure below).

One implies the other, it seems to me.

I do not agree. One changes the position of O10, while the other changes O9. 
What I would like to do is to manually make the O9 and O10 in the same plane 
with aromatic ring if possible. I was able to rotate O10 along the C07-N08 bond 
but not 09, which I shall be able to do in coot?



If I understand correctly what you are saying then that's a bug. 
C06-C07-N08-O09 defines a rotatable bond (C07-N08) about which one side of the 
fragment (the smaller side) is rotated - i.e. you cannot move O10 without 
moving O09 also. That's how it should work. It seems that you are saying that 
that is not how it's working for you.



By the way, I was able to make O9 and O10 in the same plane with the aromatic 
ring by adding planar restraint. I would like to know why I could not rotate O9 
along the N08-C07 bond if possible.

Me too. It does seem very curious. Is there a missing bond restraint - what 
happens when you display using colour-by-chain?

Paul.





To unsubscribe from the COOT list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1

This message was issued to members of www.jiscmail.ac.uk/COOT, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/


Re: [COOT] [EXTERNAL] Re: Bug in coot to change torsion angles?

2022-08-30 Thread Gong, Zhen
Hi Paul,

Thank you very much for your timely reply. There are still some puzzles for me.

I would like to change some torsion angles of a ligand in coot but it did not 
work as what I expected. I was able to play with the torsion angles formed by 
C06 C07 N08 O10 but not C06 C07 N08 O09 (shown in the figure below).

One implies the other, it seems to me.

I do not agree. One changes the position of O10, while the other changes O9. 
What I would like to do is to manually make the O9 and O10 in the same plane 
with aromatic ring if possible. I was able to rotate O10 along the C07-N08 bond 
but not 09, which I shall be able to do in coot?

By the way, I was able to make O9 and O10 in the same plane with the aromatic 
ring by adding planar restraint. I would like to know why I could not rotate O9 
along the N08-C07 bond if possible.

[cid:image001.png@01D8BCA9.3A780A30]
The restraint file for the torsion angle (C06 C07 N08 O10) is defined quite 
well: LIG CONST_23 O09 N08 C07 C06 -5.58, which is close to zero. Therefore, 
O9, O10 N08 should be more or less in the same plane with the benzene ring.



I agree.

By default Coot does not refine torsion angles.

Also torsions named "CONST" are not refined.

You can turn on torsion angle refinement in the R/RC dialog
I turned on the torsion angle refinement in the R/RC dialog and did real space 
refinement again, but O9 and O10 were still not in the same plane with the 
aromatic ring. I thought CONST means that this torsion angle is fixed (for 
example in the benzene ring), while Var means variable (flexible). This ligand 
has a cyclohexane ring also. The torsion angles for the carbons in the 
cyclohexane ring is labelled with Var_... because these angles are flexible. 
Could you please also help clarify this puzzle?

My last question is what does esd mean? Estimated standard deviation? The esd 
is zero for all the torsion angles labelled with CONST. Perhaps this is why the 
restraint was not used during phenix.refine?

I really appreciate your time and effort to help!

Best regards,
Zhen

From: Paul Emsley 
Date: Tuesday, August 30, 2022 at 19:21
To: Gong, Zhen , COOT@JISCMAIL.AC.UK 

Subject: [EXTERNAL] Re: Bug in coot to change torsion angles?


On 30/08/2022 22:26, Gong, Zhen wrote:
Dear Coot developers and users,

I would like to change some torsion angles of a ligand in coot but it did not 
work as what I expected. I was able to play with the torsion angles formed by 
C06 C07 N08 O10 but not C06 C07 N08 O09 (shown in the figure below).



One implies the other, it seems to me.




The restraint file for the torsion angle (C06 C07 N08 O10) is defined quite 
well: LIG CONST_23 O09 N08 C07 C06 -5.58, which is close to zero. Therefore, 
O9, O10 N08 should be more or less in the same plane with the benzene ring.



I agree.

By default Coot does not refine torsion angles.

Also torsions named "CONST" are not refined.

You can turn on torsion angle refinement in the R/RC dialog


If I click on the ligand with “real space refinement”, it should make the O9, 
O10 N08 more or less in the same plane with the benzene ring. However, coot did 
not do it.

If you don't have plane restraints for the nitro and ring atoms then coot 
doesn't know (hasn't been told) to keep them planar. All you've done is said to 
Coot "here's a torsion angle, don't use it.|



The corollary of the above is to turn on torsion angle refinement and rename 
the torsion var_23.

You could also add O09 to the ring plane restraint with perhaps double or 
triple the esd.



Paul.





To unsubscribe from the COOT list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1

This message was issued to members of www.jiscmail.ac.uk/COOT, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/


[COOT] Bug in coot to change torsion angles?

2022-08-30 Thread Gong, Zhen
Dear Coot developers and users,

I would like to change some torsion angles of a ligand in coot but it did not 
work as what I expected. I was able to play with the torsion angles formed by 
C06 C07 N08 O10 but not C06 C07 N08 O09 (shown in the figure below). Is it a 
bug or some geometry knowledge that I do not know?

The restraint file for the torsion angle (C06 C07 N08 O10) is defined quite 
well: LIG CONST_23 O09 N08 C07 C06 -5.58, which is close to zero. Therefore, 
O9, O10 N08 should be more or less in the same plane with the benzene ring. If 
I click on the ligand with “real space refinement”, it should make the O9, O10 
N08 more or less in the same plane with the benzene ring. However, coot did not 
do it.

Can anyone please help? Many thanks in advance!

Best,
Zhen

[cid:image002.png@01D8BC94.06712650]



To unsubscribe from the COOT list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1

This message was issued to members of www.jiscmail.ac.uk/COOT, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/


Re: [COOT] [EXTERNAL] Re: Mac Coot 0.9.8.2 crash for All-atom Refine

2022-06-13 Thread Gong, Zhen
Hi Paul,

Thank you very much for your reply. I checked my memory. I still have 12.83 GB 
available. I was able to do the “All-atom refine” in Coot 0.9.6 EL but not 
0.9.8.2. And I encountered this error when I do real space refinement with 
several residues. Therefore, I do think that the problem might be due to lack 
of restraint library for proteins.

[cid:image001.png@01D87F48.49E1D760]

Best,
Zhen
From: Paul Emsley 
Date: Monday, June 13, 2022 at 17:03
To: Zhen , COOT@JISCMAIL.AC.UK 
Subject: [EXTERNAL] Re: Mac Coot 0.9.8.2 crash for All-atom Refine

On 13/06/2022 21:35, Zhen wrote:
> I have installed on my mac Coot 0.9.8.2 (macOS Monterey version 12.4). Every 
> time when I clicked “All-atom Refine”, the coot crashed. I also encountered 
> errors when I do Real Space Refinement.


> The error was 'BadAlloc (insufficient resources for operation)'.

My guess is that you don't have enough memory to turn on all atom refinement.

Coot could (should?) do a better job at telling you that.

Paul.



To unsubscribe from the COOT list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1

This message was issued to members of www.jiscmail.ac.uk/COOT, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/