-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Tom and chimera users, I'm afraid I wasn't clear. What I would like to have is not all the asym units within a specific radius, but all the symmetry-related molecules (that may come from different asymmetric units and different unit cells) that make contact to a specific selection within a certain distance. Let's say my asymmetric unit contains two molecules A and B. If I am interested in a loop in molecule A, I would like to see the symmetry-related molecules (whether A or B, from any asymmetric unit in the crystal) that are close to this loop _and_ only these ones, i.e. not the whole asymmetric unit to which they belong. To get the positions of these molecules you need to apply both the crystal symmetry operators _and_ the lattice translations. This is what the programs we use to build our models normally do, but I understand that building a model in electronic density is something different from chimera aims. That's why I suggested to put it in the long-term wish-list ;-) Cheers, Miguel Thomas Goddard escribió:

Hi Elaine,

I agree it would be nice to add to Unit Cell. But that is a bigger project.

I do not know how many unit cells Brady wants. It could be a million atoms. The script I sent him only makes one PDB model and moves it around so the script will not run into memory limitations.

Miguel Ortiz-Lombardia sent email replying to the crystal coordinates email (but not to chimera-users, ugh) saying he wants all asym units within a specific radius. A more common need I think is a all asym units that make contact with a specified distance. So much to do and so little time.

Tom

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

- -- Miguel Ortiz Lombardía Centro de Investigaciones Oncológicas C/ Melchor Fernández Almagro, 3 28029 Madrid, Spain Tel. +34 912 246 900 Fax. +34 912 246 976 email: molatwork@yahoo.es www: http://www.ysbl.york.ac.uk/~mol/ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Le travail est ce que l'homme a trouvé de mieux pour ne rien faire de sa vie. (Raoul Vaneigem) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (GNU/Linux) iD8DBQFFrd8fF6oOrDvhbQIRAjGgAJ4tbLdax+f3KunzVUvpjguSp8OgXACbBY8f qdpOfQvVELxiL8hw1oOTdy0= =epnT -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Tom, Thanks to you! This is an awesome tool!! Let me just add a comment from someone working in protein/nucleic acids (not virus) crystallography: Although your use of the term "NCS" though very correct, it may be confusing for some people in our field. The reason is that 90% (?) of the time we deposit a full asymmetric unit, regardless of whether there is or not non-crystallographic symmetry. Therefore, coordinates for a full AU still can represent a crystal where _there_is_ NCS. This is because we normally refine all atoms in the AU except for the 10% (?) of the time when they are constrained (NCS-related molecules in the AU are forced to be in perfectly equivalent positions). We normally use restraints more or less relaxed or even no restraints at all if resolution is high enough. Only for very low resolution crystals with constrained NCS, people deposit the "independent" part of the AU and give MATRIX records for the rest of the positions. So, my suggestion is that you might substitute "strict NCS" (or "explicit NCS", etc) for "NCS" and make clear that NCS may be present even if the coordinates contain the full AU. But perhaps your users are mainly virus crystallographers or EM people, who wouldn't care much... So, just a suggestion! Would it be very difficult to change the code so each chain in the AU is represented by a green sphere and the distance calculations are made relative to each of them? Also, in the other AU's, each sphere would represent a chain... As you see, this relates to my suggestion in a way, though the different chains in the AU need not to be related by NCS. I've been looking in the code but haven't failed to find where the lists are generated... Thank you again! Even as it is, this tool is really helpful to analyse molecular contacts! Best regards, Miguel Thomas Goddard escribió:

Hi Miguel,

Thanks for the clarification. I agree that it is important to get just the contacting chains, not the whole asymmetric units, and that both crystal symmetries and unit cell translations need to be considered.

The basic code to achieve this is already in Chimera so it should not be difficult to add. The needed routines are used by the "crystal contacts" experimental plugin

http://www.cgl.ucsf.edu/chimera/experimental/crystal_contacts/clashes.html

although that tool works at the level of whole asymmetric units. Some adaptations of that code would allow it to instead consider individual chains and display all atoms of each chains instead of a single sphere for each chain. I'll let you know when I have written code you can try.

Tom

- -- Miguel Ortiz Lombardía Centro de Investigaciones Oncológicas C/ Melchor Fernández Almagro, 3 28029 Madrid, Spain Tel. +34 912 246 900 Fax. +34 912 246 976 email: molatwork@yahoo.es www: http://www.ysbl.york.ac.uk/~mol/ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Le travail est ce que l'homme a trouvé de mieux pour ne rien faire de sa vie. (Raoul Vaneigem) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2.2 (GNU/Linux) iD8DBQFFr0pYF6oOrDvhbQIRAihzAJ4nyNYsU+rvHMXv3MxUqhazw16/HQCeNk7+ qWgtrurtkVMKQHDo6rWrpJ4= =oxrA -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Tom, I agree that the first tool would be better. But I liked the spheres representation for having a quick idea of the symmetry-related molecules interacting. It's also extremely useful for pedagogical purposes. The proportionality between the number of contacts and the radii of the cylinders is such a clever idea! I will try to make a version where the maximum radius will be assigned to the maximum number of contacts so the differences are enhanced for crystals of proteins with a lot less contacts than proteins in virus capsides. I'll be waiting for that new tool, but I can understand it's not on the top of the list ;-) Thank you for your help! Best regards, Miguel En/na Thomas Goddard ha escrit:

Hi Miguel,

I agree that the term non-crystallographic symmetry (NCS) can be confusing. The crystal contacts Chimera tool was developed specifically for fixing the MTRIX records in virus entries in the PDB so the description uses terminology that makes sense for virus crystal structures.

I think the crystal contacts tool could be adapted so that the spheres represent individual PDB chains. This would be similar to your earlier request for a tool that just displays contacting chains. But I think the tool you described earlier would be better (only displays contacting chains from entire crystal, shows them as atomic models) and is just as easy to implement. Not sure if I will have time to work on it in coming weeks or months -- I haven't yet scheduled it into my long list of projects.

Tom

- -- Miguel Ortiz Lombardía Centro de Investigaciones Oncológicas C/ Melchor Fernández Almagro, 3 28029 Madrid, Spain Tel. +34 912 246 900 Fax. +34 912 246 976 e-mail: molatwork@yahoo.es - ---------------------------------------------------------------------- Et ainsi ne pouvant faire que ce qui est juste fût fort, on a fait que ce qui est fort fût juste. Blaise Pascal, Pensées -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (Darwin) iD8DBQFFr+Z1F6oOrDvhbQIRAvQpAJ9oOUH/ZtAOIJcBFSd3NBrAkJhWOQCfQhLo WmGfaMv1g8owSPEK8TDmmiQ= =bzzR -----END PGP SIGNATURE-----