Hi Smith,

  The Unit Cell tool shows 6 copies of the 2ZAM asymmetric unit.  Since the crystal symmetry extends to infinity Chimera has to decide which 6 to show.  How does it decide?  There is nothing smart about it — it simply takes the 6 space group symmetry operators.  Here they are for space group P65 taken from the 2ZAM.pdb header

REMARK 290 CRYSTALLOGRAPHIC SYMMETRY                                            
REMARK 290 SYMMETRY OPERATORS FOR SPACE GROUP: P 65                             
REMARK 290                                                                      
REMARK 290      SYMOP   SYMMETRY                                                
REMARK 290     NNNMMM   OPERATOR                                                
REMARK 290       1555   X,Y,Z                                                   
REMARK 290       2555   -Y,X-Y,Z+2/3                                            
REMARK 290       3555   -X+Y,-X,Z+1/3                                           
REMARK 290       4555   -X,-Y,Z+1/2                                             
REMARK 290       5555   Y,-X+Y,Z+1/6                                            
REMARK 290       6555   X-Y,X,Z+5/6                                             
REMARK 290                                                                      
REMARK 290     WHERE NNN -> OPERATOR NUMBER                                     
REMARK 290           MMM -> TRANSLATION VECTOR                                  

It applies those 6 translations and reflections (they are shown above in coordinates where unit cell size is 1,1,1).  When you apply those 6 it doesn’t necessarily give you the closest packed 6 copies of your molecule.  That is where the “pack” option in the Unit Cell dialog comes in.  With “pack” enabled, after using those symmetry operators Chimera will shift each copy by one or more unit cells along each of the 3 axes so that the molecule center lies in the unit cell outline box that is displayed.  If in the coordinates with unit cell size 1,1,1, the center of a molecule copy lies outside 0-1,0-1,0-1 then it shifts the coordinates by an integer along each axis so it lies inside.  Such shifts just show the copy in the adjoining unit cell.  You might want other choices of which 6 copies to show.  To achieve that you could show 2 by 2 by 2 unit cells (48 = 6*2*2*2 copies) then simply delete the copies you don’t want.

Tom


On Jan 31, 2017, at 10:19 AM, Elaine Meng <meng@cgl.ucsf.edu> wrote:


Hi Smith,
The Unit Cell results for a structure with and without “pack” are equivalent because the symmetry describes a repeating pattern: each single copy could be shown in any of several places around the first copy.  In other words, as I said before, there are multiple equally correct ways of presenting the same information. 

Although I cannot spend a lot of time looking at your specific data, I see the 3 structures are all the same space group with similar unit cell dimensions and angles, and the paper says “we could present nucleotide-induced conformational changes in the same crystal lattice by eliminating possible artificial structural differences because of different crystal packing.”  Therefore I infer that the crystal packing is approximately the same in the 3 structures.  It cannot be EXACTLY the same, however, since there are conformational changes between the structures.

Since I didn’t write the Unit Cell code or look at the symmetry matrices in detail, I don’t know why the structures don’t give the same arrangement without “pack,” but I don’t think it is important, since the different possible arrangements for a single structure are equally correct.  I hope this helps,
Elaine

P.S. please send questions to chimera-users rather than to me individually - thanks

Begin forwarded message:

From: "Smith Liu" <smith_liu123@163.com>
Subject: Fw:Re: [Chimera-users] on chimera unit cell function
Date: January 31, 2017 at 7:01:54 AM PST

Dear Elaine,

Related to our previous communication on "on chimera unit cell function" , here I have more to discuss with you.

Attached 1 was the paper “Nucleotide-Dependent Conformational Changes and
Assembly of the AAA ATPase SKD1/VPS4B” published in Traffic 2008. PDB ID 2ZAM, 2ZAN, 2ZAO were 3 pdbs from this paper, and 2ZAM was the VPS4B without any nucleotide, 2ZAN was the VPS4B bound with ATP, 2ZAO as the VPS4B bound with ADP.
 
As the paper wrote, “Because our attempts to crystallize SKD1 (VPS4B) in the presence
of ATP or ADP failed, we soaked the apo form crystals in solutions of ATP or ADP”, thus the 2ZAN and 2ZAO were got by soaking the crystal for 2ZAM with ATP and ADP separately. All 3 crystals were with space group P 65.
 
The pdb for the 6 mates for each of 2ZAM, 2ZAN and 2ZAO are attached in attachment 2, 3, 4 separately (prepared without clicking ""Pack molecules in unit cell"). If viewed by Chimera, you will find the mates of 2ZAM arrange in the same way as the mates of 2ZAO, i.e., the 6 mates connect each other closely (side by side) and packed in a screw way.  However for the 6 mates of 2ZAN, although still in screw way, the 6 mates arrange not closely but have a space between the adjacent mates. Correspondingly, the diameters of the z-axis projection of the 6 mates of 2ZAM and 2ZAO were much smaller than that of the 6 mates of 2ZAN.

However, if we by Chimera we click the "Pack molecules in unit cell",we will find the 6 mates of 2ZAN displayed exactly (or very similarly) as the 6 mates of 2ZAM and 2ZAO (Am I right).

Then my question would be, in the crystal for 2ZAN and 2ZAO, do the proteins arranged in exactly the same pattern? And why if we did not click ""Pack molecules in unit cell", the mates of 2ZAN and 2ZAO were displayed differently, and with clicking ""Pack molecules in unit cell",the mates of 2ZAN and 2ZAO were displayed similarly?

If the protein arranged differently in the crystal for 2ZAN and 2ZAO, then in the crystal packing of 2ZAN, how can I identify the 6 protein which were from the 6 screwed mates in 2ZAM?

I am looking forward to getting a reply from you.
Smith
-------- Forwarding messages --------
From: "Elaine Meng" <meng@cgl.ucsf.edu>
Date: 2017-01-30 04:05:22
To:  "Smith Liu" <smith_liu123@163.com>
Cc:  "chimera-users@cgl.ucsf.edu BB" <chimera-users@cgl.ucsf.edu>
Subject: Re: [Chimera-users] on chimera unit cell function
Hi Smith,
The number of copies and their positions are all correct no matter whether you choose the “pack” option or not.  If you cut out a part of the real crystal, there will be a real copy in the same place as the copy from Chimera.  So imagine if you had specified an infinite (or large) number of unit cells in Chimera, a chunk from the middle of the result would look the same as some chunk of the real idealized crystal, regardless of whether you used “pack”.  However, since you are probably specifying 1 x 1 x 1 or some small number of unit cells, yes choose the “pack” option to show the copies packed together, instead of with some spaces where a copy is shown inside a neighboring box rather than in the central outline box.
Best,
Elaine

> On Jan 29, 2017, at 5:10 AM, Smith Liu <smith_liu123@163.com> wrote:
> 
> Dear Elaine,
> 
> I am still confusing on the function of "Pack molecules in unit cell" in Chimera. If we cut out a cube (or a volume) of crystal based on which a pdb was got, do you think the molecules packed in the cube in the way without "Pack molecules in unit cell" clicked, or the molecules packed in the cube in the way with "Pack molecules in unit cell" clicked?
> 
> I am looking forward to getting a reply from you.
> 
> Smith
> 
> 
> At 2017-01-29 00:50:44, "Elaine Meng" <meng@cgl.ucsf.edu> wrote:
> >Hi Smith,
> >It is my understanding that you will get the same number of copies whether or not you choose the “pack” option, and that there are multiple equally correct ways in which the copies will be positioned according to the symmetry information.  The “pack” option simply chooses, out of these many equally correct alternatives, a configuration that places the centers of all the copies inside one unit cell outline box instead of spreading them out among neighboring boxes.
> >I hope this helps,
> >Elaine
> >-----
> >Elaine C. Meng, Ph.D.                       
> >UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
> >Department of Pharmaceutical Chemistry
> >University of California, San Francisco
> > 
> >> On Jan 28, 2017, at 1:47 AM, Smith Liu <smith_liu123@163.com> wrote:
> >> 
> >> Dear All,
> >> 
> >> In Chimera, we do Tools →Higher-Order Structure→Unit Cell, then what will be the diffenrence if we clicked "Pack molecules in unit cell" in comparison with if we did not click "Pack molecules in unit cell"? Does it mean that what we observed (or what Chimera displayed) if we click "Pack molecules in unit cell" , was really part of what it displayed if we did not click "Pack molecules in unit cell" but we increased the "Number of Cells" significantly?
> >> 
> >> I am looking forward to getting a reply from you.
> >> 
> >> Smith

_______________________________________________
Chimera-users mailing list: Chimera-users@cgl.ucsf.edu
Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users