This is something you would normally not do in the course of refining a protein structure, because the structure would no longer correspond to the observed structure factors and no longer be consistent with the symops of the particular space group.
There are applications for it in multi-crystal averaging, or to compare directly the density map for structures from non-isomorphous crystals. For doing this it is important to understand the difference between a MAP file (.map, .ccp4) and a DATA file (.mtz). There are programs for rotating/translating electron density (from a map file) given the rotation-translation operator. I think ccp4 mapmask(?), also the uppsala RAVE and MAVE packages.
So you could generate a map from your mtz (ccp4 prog fft), get the operator relating new position to old
(coot or LSQAB? or LSQMAN), transform ("skew") the map to the new location in a p1 cell, and calculate new p1 phased structure factors with SFALL (or use the map file directly in coot). You need a mask for the skew operation, you can make it from your coordinates using mapmask or RAVE mama.
On 12/18/2017 10:40 AM, Smith Liu wrote:
> sorry, how i move the mtz into the transformed pdb for the question in my previous email?
>
>
>
>
> Smith Liu
> 邮箱:[log in to unmask]
>
> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?iconUrl=https://nos.netease.com/mail-online/qiyelogo/defaultAvatar.png&name=Smith Liu&uid=smith_liu123%40163.com&ftlId=1&items=%5B%22%E9%82%AE%E7%AE%B1%EF%BC%9Asmith_liu123%40163.com%22%5D>
>
> 签名由 网易邮箱大师 <https://mail.163.com/dashi/dlpro.html?from=mail88> 定制
>
>
>
> 在2017年12月18日 23:37,Smith Liu <mailto:[log in to unmask]> 写道:
> thanks. i may mean something other. for example, if i rotate the pdb by 30 degree (or 29.5 degree), or i shift the pdb along x-axis by something for example 0.123*a, then how i move the mtz map correspondingly for the fitting of mtz into the transformed map?
>
>
>
>
> Smith Liu
> 邮箱:[log in to unmask]
>
> <https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?iconUrl=https://nos.netease.com/mail-online/qiyelogo/defaultAvatar.png&name=Smith Liu&uid=smith_liu123%40163.com&ftlId=1&items=%5B%22%E9%82%AE%E7%AE%B1%EF%BC%9Asmith_liu123%40163.com%22%5D>
>
> 签名由 网易邮箱大师 <https://mail.163.com/dashi/dlpro.html?from=mail88> 定制
>
>
>
> 在2017年12月18日 21:58,[log in to unmask] <mailto:[log in to unmask]> 写道:
>
> Dear Smith,
>
> The map extends through the whole crystal. What happens is that the map is calculated around the atom you clicked on during centering. So by centering on your transformed pdb, you will sample the same map at a different position. Just load your transformed pdb and untransformed mtz and try.
>
> If the transformed pdb does not fit the map, something went wrong during the transformation of your pdb. If you have applied an origin shift (is not equal to applying a crystallographic symmetry operation), you have to recalculate the mtz, e.g. by running another round of refinement.
>
> I hope this is clear so.
>
> Herman
>
> *Von:*Smith Liu [mailto:[log in to unmask] <mailto:[log in to unmask]>]
> *Gesendet:* Montag, 18. Dezember 2017 14:52
> *An:*Schreuder, Herman /DE
> *Betreff:* [EXTERNAL] Re: [ccp4bb] AW: Re: [ccp4bb] coordinate transformation
>
> you mean the mtz map will transform simutaneously?
>
>
>
>
> *Smith Liu*
>
> 邮箱:[log in to unmask] <mailto:[log in to unmask]>
>
> 签名由网易邮箱大师 <https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.163.com_dashi_dlpro.html-3Ffrom-3Dmail88&d=DwMGbw&c=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo&r=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ&m=VsiJhHbT-qp6n3QHdvilZeqY-0tA4mSqlVkx6nStzhM&s=0Zu9ZB12d-Kh3exqmmGcW7PZIzCGKTXlXat7Qffx-lk&e=> 定制
>
>
>
>
> 在2017年12月18日 21:26,[log in to unmask] <mailto:[log in to unmask]> 写道:
>
> If you use coot with on the fly map calculation (e.g. you load an mtz and not a map file), you do not need to transform the map. Otherwise I would recommend to run one more round of refinement and produce a new map your usual way. This will also get rid of any rounding errors due to the transformation.
>
> Best,
>
> Herman
>
> *Von:*CCP4 bulletin board [mailto:[log in to unmask] <mailto:[log in to unmask]>] *Im Auftrag von *Smith Liu
> *Gesendet:* Montag, 18. Dezember 2017 14:16
> *An:* [log in to unmask] <mailto:[log in to unmask]>
> *Betreff:* [EXTERNAL] Re: [ccp4bb] coordinate transformation
>
> Dear All,
>
> If I have a set of PDB with the corresponding density map, after I transform the PDB based on the suggestion of everybody, is any way to transform the map so that the map will be fit with the transformed PDB?
>
> Smith
>
>
>
>
> At 2017-12-18 18:39:34, "Eleanor Dodson" <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> I showed you pdbset ..
>
> Find the centre of mass for your assembly.
>
> Move it where you will
>
> pdbset xyzin mow.pdb
>
> end
>
> Find CoM 0.7 1.3 -0.2
>
> Hmm - a little thought - centre at 1 -1 0 say
>
> pdbset yzin now.pdb xyzout changed.pdb
>
> symgen x , y-2, z
>
> end
>
> New CoM 0.7 -0.7 -0.2
>
> Eleanor
>
> On 18 December 2017 at 00:19, Edward A. Berry <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
> Neat idea!
> And do you have a 1-line command for setting all the coordinates to 1,1,1? or 0.1,0.1,0.1 if I still want it near the origin but biased toward the inside of the positive-going cell?
> eab
>
>
>
> On 12/14/2017 07:23 PM, James Holton wrote:
>
> What I usually do for this is make a copy of the PDB file and change all the atom x-y-z positions to "1.000". Then I use something like reforigin or my "origins.com <https://urldefense.proofpoint.com/v2/url?u=http-3A__origins.com&d=DwMGbg&c=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo&r=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ&m=kr4BAC6CGF8MkSeqVDxlDaQ3WprGVrjotZPIWrNrdts&s=RFDcNoPw6eYR8Aka4k3PWnf4_OIc98ZgWFd-LYwSoHo&e=>" script to shift the original coordinates via allowed symmetry operations, origin shifts, or perhaps indexing ambiguities until it is as close as possible to the "reference", which is at 1,1,1. I use 1,1,1 instead of 0,0,0 because there are generally at least two symmetry-equivalent places that are equidistant from the origin. Declaring the reference to be a bit off-center breaks that ambiguity, and also biases the result toward having all-positive x,y,z values.
>
>
> In case it is interesting, my script is here:
>
> http://bl831.als.lbl.gov/~jamesh/scripts/origins.com <https://urldefense.proofpoint.com/v2/url?u=http-3A__bl831.als.lbl.gov_-7Ejamesh_scripts_origins.com&d=DwMGbg&c=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo&r=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ&m=kr4BAC6CGF8MkSeqVDxlDaQ3WprGVrjotZPIWrNrdts&s=9dlreI0nRWqq1Feor3gq_OOcpIxVPjpdRl2KTg9fMqg&e=>
>
>
> You need to have the CCP4 suite set up for it to work. Run it with no arguments to get instructions.
>
>
> -James Holton
>
> MAD Scientist
>
>
> On 12/13/2017 5:50 AM, Kajander, Tommi A wrote:
>
>
> Hello,
>
> If someone could point this out would be very helpful... Wasnt there a simple script somewhere that would transfer coordinates close to origin - if they for some reason are not? Just cant find anything right away. Sure i have done this before...
>
>
> Thanks,
>
> Tommi
>
|