Re: [ccp4bb] difficult molecular replacement solution

[Eleanor Dodson]

Well, trying to solve a MR structure with low homology (20%) and several
molecules is always a challenge...
Use any information you have.
How many molecules are possible/probable given the mol.wt and unit cell?
(Matthews coeff gives suggestions..) But remember solvent content can vary
from 30% to 70% and more.

 Is the structure likely to form dimers? trimers? etc. Biochemistry can
help, and the self rotation can also give clues, although it can be a pain
to interpret..

And if so is there a dimer model or trimer or whatever? May be better to
search for fewer copies of the oligmer..

But if you get some sort of reasonable solution, then start refinement and
rebuilding - if you can improve a single copy then you can use that as a
new and better search model..


I would expect bigger increases in the LLG and TGscore if the 5th and 6th
molecule were correct but maybe not..
You know you have solved it when the rebuilding bites..
\Good luck Eleanor



On Wed, 13 Nov 2019 at 17:42, Tanner, John J.  wrote:

> Dear Rob,
>
> Based on our experience with difficult MR cases, I recommend performing
> refinement on the coordinates from MR (I prefer PHENIX simulated annealing
> for this step). Then send the refined map - WITHOUT the model - to
> automated building with density modification. The auto-built model may give
> you an indication of how many chains are in the asymmetric unit. You could
> do this procedure for each of the MR solutions (N=4,5,6) and compare the
> results. Also, if you enable automatic detection of NCS during DM, you may
> find that 6 NCS operators are found even when N=4 or 5; this would give you
> confidence that N=6 is possible.
>
> Also look at the cross rotation function to see how many strong peaks are
> present. I’m sure the peaks are listed in Phaser, but it is easier for me
> to find them in the MOLREP output.  MOLREP also provides a matrix that
> tells you which cross RF peaks are related by symmetry and which are
> unique, which is helpful.
>
> And David Schuller mentioned inspecting the self-RF, which is a good idea.
>
> Finally, do you know anything about the oligomeric state of the protein?
> If it is a pentamer, N=6 seems unlikely. If it is a dimer, N=5 in P212121
> seems unlikely.
>
> Jack
>
> John J. Tanner
> Professor of Biochemistry and Chemistry
> Associate Chair of Biochemistry
> Department of Biochemistry
> University of Missouri
> 117 Schweitzer Hall
> 503 S College Avenue
> Columbia, MO 65211
> Phone: 573-884-1280
> Email: tanne...@missouri.edu
> http://faculty.missouri.edu/~tannerjj/tannergroup/tanner.html
> Lab: Schlundt Annex rooms 3,6,9, 203B, 203C
> Office: Schlundt Annex 203A
>
> On Nov 13, 2019, at 8:33 AM, Robert S Phillips  wrote:
>
> I have been working on a protein structure which has been hard to solve by
> molecular replacement.
>
> Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
> Space group: P 21 21 21
>
> The problem is that the homologues have only ~20% identity, and there are
> multiple chains in the asymmetric unit.  The question is how many.  It
> could be 4, 5, or 6 chains.
>
> N  solvent   P
>  4  0.602   3.090.225
>  5  0.502   2.470.388
>  6  0.403   2.060.229
>
> I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all
> possible space groups, and P212121 was the best solution.  These are the
> results.
>
> N  LLG   TFZ
> 4  104.97.5
> 5  137.57.7
> 6  166.28.3
>
>  Am I correct to conclude that there are 6 chains in the asymmetric unit?
>
> Rob
>
> Robert S. Phillips
> Professor of Chemistry and of Biochemistry and Molecular Biology
> University of Georgia
> Athens, GA 30602
> Phone: (706) 542-1996
> Fax: (706) 542-9454
> E-mail: rsphill...@chem.uga.edu
> Web:  http://tryptophan.net
> 
>
> --
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>
>
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] difficult molecular replacement solution

[Jon Cooper]

I am interested to know the resolution on the basis that the better it diffracts, the lower the solvent content, generally. Have you tried Bernard Rupp's server which gives you probabilities for the No mols per AU, partly based on resolution? The LLGs look challenging - I've seen higher ones refine with only random R-free values. On 13 Nov 2019 17:42, "Tanner, John J."  wrote:
Dear Rob,


Based on our experience with difficult MR cases, I recommend performing refinement on the coordinates from MR (I prefer PHENIX simulated annealing for this step). Then send the refined map - WITHOUT the model - to automated building with density
 modification. The auto-built model may give you an indication of how many chains are in the asymmetric unit. You could do this procedure for each of the MR solutions (N=4,5,6) and compare the results. Also, if you enable automatic detection of NCS during DM,
 you may find that 6 NCS operators are found even when N=4 or 5; this would give you confidence that N=6 is possible. 


Also look at the cross rotation function to see how many strong peaks are present. I’m sure the peaks are listed in Phaser, but it is easier for me to find them in the MOLREP output.  MOLREP also provides a matrix that tells you which cross RF
 peaks are related by symmetry and which are unique, which is helpful. 


And David Schuller mentioned inspecting the self-RF, which is a good idea.


Finally, do you know anything about the oligomeric state of the protein? If it is a pentamer, N=6 seems unlikely. If it is a dimer, N=5 in P212121 seems unlikely. 


Jack 
















John J. Tanner

Professor of Biochemistry and Chemistry

Associate Chair of Biochemistry


Department of Biochemistry

University of Missouri
117 Schweitzer Hall
503 S College Avenue
Columbia, MO 65211
Phone: 573-884-1280

Email: tannerjj@missouri.edu
http://faculty.missouri.edu/~tannerjj/tannergroup/tanner.html

Lab: Schlundt Annex rooms 3,6,9, 203B, 203C

Office: Schlundt Annex 203A















On Nov 13, 2019, at 8:33 AM, Robert S Phillips  wrote:



I have been working on a protein structure which has been hard to solve by molecular replacement.  





Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
Space group: P 21 21 21





The problem is that the homologues have only ~20% identity, and there are multiple chains in the asymmetric unit.  The question is how many.  It could be 4, 5, or 6 chains.




N  solvent   P


 4      0.602           3.09            0.225             

 5      0.502           2.47            0.388        

 6      0.403           2.06            0.229              





I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all possible space groups, and P212121 was the best solution.  These are the results.




N  LLG   TFZ


4          104.9        7.5

5          137.5        7.7

6          166.2        8.3




 Am I correct to conclude that there are 6 chains in the asymmetric unit?




Rob





Robert S. Phillips
Professor of Chemistry and of Biochemistry and Molecular Biology 

University of Georgia 
Athens, GA 30602
Phone: (706) 542-1996
Fax: (706) 542-9454
E-mail: rsphillips@chem.uga.edu
Web:  http://tryptophan.net







To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1








To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] difficult molecular replacement solution

[Phoebe A. Rice]

I like to try molrep with something missing from the search model so that I can 
see if the resulting map includes density for something that I know should be 
there but the computer doesn’t.  As others have pointed out, it also bolsters 
believability if the result agrees with a self-rotation function and shows nice 
packing.

~~~
Phoebe A. Rice
Dept. of Biochem & Mol. Biol. and
  Committee on Microbiology
https://voices.uchicago.edu/phoebericelab/


From: CCP4 bulletin board  on behalf of Robert S 
Phillips 
Reply-To: Robert S Phillips 
Date: Wednesday, November 13, 2019 at 8:33 AM
To: "CCP4BB@JISCMAIL.AC.UK" 
Subject: [ccp4bb] difficult molecular replacement solution

I have been working on a protein structure which has been hard to solve by 
molecular replacement.

Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
Space group: P 21 21 21

The problem is that the homologues have only ~20% identity, and there are 
multiple chains in the asymmetric unit.  The question is how many.  It could be 
4, 5, or 6 chains.

N  solvent   P
 4  0.602   3.090.225
 5  0.502   2.470.388
 6  0.403   2.060.229

I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all possible 
space groups, and P212121 was the best solution.  These are the results.

N  LLG   TFZ
4  104.97.5
5  137.57.7
6  166.28.3

 Am I correct to conclude that there are 6 chains in the asymmetric unit?

Rob

Robert S. Phillips
Professor of Chemistry and of Biochemistry and Molecular Biology
University of Georgia
Athens, GA 30602
Phone: (706) 542-1996
Fax: (706) 542-9454
E-mail: rsphill...@chem.uga.edu
Web:  
http://tryptophan.net<https://pod51004.outlook.com/owa/redir.aspx?C=ccbf42ffea5f48b1bf8e9bb950454bab=http%3a%2f%2ftryptophan.net>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] difficult molecular replacement solution

[Tanner, John J.]

Dear Rob,

Based on our experience with difficult MR cases, I recommend performing 
refinement on the coordinates from MR (I prefer PHENIX simulated annealing for 
this step). Then send the refined map - WITHOUT the model - to automated 
building with density modification. The auto-built model may give you an 
indication of how many chains are in the asymmetric unit. You could do this 
procedure for each of the MR solutions (N=4,5,6) and compare the results. Also, 
if you enable automatic detection of NCS during DM, you may find that 6 NCS 
operators are found even when N=4 or 5; this would give you confidence that N=6 
is possible.

Also look at the cross rotation function to see how many strong peaks are 
present. I’m sure the peaks are listed in Phaser, but it is easier for me to 
find them in the MOLREP output.  MOLREP also provides a matrix that tells you 
which cross RF peaks are related by symmetry and which are unique, which is 
helpful.

And David Schuller mentioned inspecting the self-RF, which is a good idea.

Finally, do you know anything about the oligomeric state of the protein? If it 
is a pentamer, N=6 seems unlikely. If it is a dimer, N=5 in P212121 seems 
unlikely.

Jack

John J. Tanner
Professor of Biochemistry and Chemistry
Associate Chair of Biochemistry
Department of Biochemistry
University of Missouri
117 Schweitzer Hall
503 S College Avenue
Columbia, MO 65211
Phone: 573-884-1280
Email: tanne...@missouri.edu
http://faculty.missouri.edu/~tannerjj/tannergroup/tanner.html
Lab: Schlundt Annex rooms 3,6,9, 203B, 203C
Office: Schlundt Annex 203A

On Nov 13, 2019, at 8:33 AM, Robert S Phillips 
mailto:p...@uga.edu>> wrote:

I have been working on a protein structure which has been hard to solve by 
molecular replacement.

Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
Space group: P 21 21 21

The problem is that the homologues have only ~20% identity, and there are 
multiple chains in the asymmetric unit.  The question is how many.  It could be 
4, 5, or 6 chains.

N  solvent   P
 4  0.602   3.090.225
 5  0.502   2.470.388
 6  0.403   2.060.229

I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all possible 
space groups, and P212121 was the best solution.  These are the results.

N  LLG   TFZ
4  104.97.5
5  137.57.7
6  166.28.3

 Am I correct to conclude that there are 6 chains in the asymmetric unit?

Rob

Robert S. Phillips
Professor of Chemistry and of Biochemistry and Molecular Biology
University of Georgia
Athens, GA 30602
Phone: (706) 542-1996
Fax: (706) 542-9454
E-mail: rsphill...@chem.uga.edu
Web:  
http://tryptophan.net


To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] difficult molecular replacement solution

[David J. Schuller]

Have you looked at a self-rotation function? Those can sometimes be very useful 
in deciphering multiplicity.


On 2019-11-13 09:33, Robert S Phillips wrote:
I have been working on a protein structure which has been hard to solve by 
molecular replacement.

Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
Space group: P 21 21 21

The problem is that the homologues have only ~20% identity, and there are 
multiple chains in the asymmetric unit.  The question is how many.  It could be 
4, 5, or 6 chains.

N  solvent   P
 4  0.602   3.090.225
 5  0.502   2.470.388
 6  0.403   2.060.229

I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all possible 
space groups, and P212121 was the best solution.  These are the results.

N  LLG   TFZ
4  104.97.5
5  137.57.7
6  166.28.3

 Am I correct to conclude that there are 6 chains in the asymmetric unit?

Rob

Robert S. Phillips
Professor of Chemistry and of Biochemistry and Molecular Biology
University of Georgia
Athens, GA 30602
Phone: (706) 542-1996
Fax: (706) 542-9454
E-mail: rsphill...@chem.uga.edu
Web:  
http://tryptophan.net



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1


--
===
All Things Serve the Beam
===
   David J. Schuller
   modern man in a post-modern world
   MacCHESS, Cornell University
   schul...@cornell.edu



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] difficult molecular replacement solution

[Kay Diederichs]

Hi Rob,

the answer to the question "how many molecules are in the ASU?" is only 
available after satisfactory model building and refinement. The Phaser results 
that you show indeed suggest 6 mols/ASU, but they don't prove it. If I were 
you, I'd give the 6-molecule solution the highest priority, and start to build 
and refine. But be prepared to revise your hypothesis if you find that some of 
the molecules overlap with others.

best,
Kay

On Wed, 13 Nov 2019 14:33:04 +, Robert S Phillips  wrote:

>I have been working on a protein structure which has been hard to solve by 
>molecular replacement.
>
>Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
>Space group: P 21 21 21
>
>The problem is that the homologues have only ~20% identity, and there are 
>multiple chains in the asymmetric unit.  The question is how many.  It could 
>be 4, 5, or 6 chains.
>
>N  solvent   P
> 4  0.602   3.090.225
> 5  0.502   2.470.388
> 6  0.403   2.060.229
>
>I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all possible 
>space groups, and P212121 was the best solution.  These are the results.
>
>N  LLG   TFZ
>4  104.97.5
>5  137.57.7
>6  166.28.3
>
> Am I correct to conclude that there are 6 chains in the asymmetric unit?
>
>Rob
>
>Robert S. Phillips
>Professor of Chemistry and of Biochemistry and Molecular Biology
>University of Georgia
>Athens, GA 30602
>Phone: (706) 542-1996
>Fax: (706) 542-9454
>E-mail: rsphill...@chem.uga.edu
>Web:  
>http://tryptophan.net
>
>
>
>To unsubscribe from the CCP4BB list, click the following link:
>https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] difficult molecular replacement solution

[Roger Rowlett]

 Look at the chain packing in your solution. If the solution is grossly
correct you will notice nice solvent channels and no glaring chain-size
holes and no gross overlaps. The initial maps should look sensible, if
noisy.

If the protein is composed of likely dimers, try a search with dimer units
to simplify the search.

Good luck.

__
Roger Rowlett

On Wed, Nov 13, 2019, 10:33 AM Robert S Phillips  wrote:

> I have been working on a protein structure which has been hard to solve by
> molecular replacement.
>
> Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
> Space group: P 21 21 21
>
> The problem is that the homologues have only ~20% identity, and there are
> multiple chains in the asymmetric unit.  The question is how many.  It
> could be 4, 5, or 6 chains.
>
> N  solvent   P
>  4  0.602   3.090.225
>  5  0.502   2.470.388
>  6  0.403   2.060.229
>
> I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all
> possible space groups, and P212121 was the best solution.  These are the
> results.
>
> N  LLG   TFZ
> 4  104.97.5
> 5  137.57.7
> 6  166.28.3
>
>  Am I correct to conclude that there are 6 chains in the asymmetric unit?
>
> Rob
>
> Robert S. Phillips
> Professor of Chemistry and of Biochemistry and Molecular Biology
> University of Georgia
> Athens, GA 30602
> Phone: (706) 542-1996
> Fax: (706) 542-9454
> E-mail: rsphill...@chem.uga.edu
> Web:  http://tryptophan.net
> 
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] difficult molecular replacement solution

[Robert S Phillips]

I have been working on a protein structure which has been hard to solve by 
molecular replacement.

Unit cell: (60.6, 172.34, 196.42, 90, 90, 90)
Space group: P 21 21 21

The problem is that the homologues have only ~20% identity, and there are 
multiple chains in the asymmetric unit.  The question is how many.  It could be 
4, 5, or 6 chains.

N  solvent   P
 4  0.602   3.090.225
 5  0.502   2.470.388
 6  0.403   2.060.229

I have run PHASER with 4, 5 and 6 chains.  I allowed it to search all possible 
space groups, and P212121 was the best solution.  These are the results.

N  LLG   TFZ
4  104.97.5
5  137.57.7
6  166.28.3

 Am I correct to conclude that there are 6 chains in the asymmetric unit?

Rob

Robert S. Phillips
Professor of Chemistry and of Biochemistry and Molecular Biology
University of Georgia
Athens, GA 30602
Phone: (706) 542-1996
Fax: (706) 542-9454
E-mail: rsphill...@chem.uga.edu
Web:  
http://tryptophan.net



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1