Dear Elaine,

I have been using measure distance a lot to get the exact distance between the markers of my interest, and cannot praise the function enough.

However, what exactly I am trying to search is whether there exist a model-model gap of specific distance that I can specify in Å..

I am currently working on a project where I placed a cylindrical model (which I open by opening a script that includes two markers to be the both ends of the filament (or simply put, marker-marker distance = actin filament length), and a Link, shown with radius of Actin filament to depict the space Actin filament would actually occupy in 3D space.

And, knowing that various form of Actin-Acrin Cross-Linking Proteins have very well defined and unique lengths, I wanted to just locate and visualize any and all region(s) of my active models (=which are solely actin filaments) that are spaced "exactly by cross-linking protein's length away from each other". Then I would be able to pin-point presence of, and location of, and potentially number of any cross-linking protein that is present.

And for Tom's script; I think it actually places n number of additional markers along the Link (or Bond) that I have selected, where additional markers share identical model ID as the model ID of the Link. Just discovered that the script doesn't work simply by selecting two end markers, but only by selecting a Link.

And here is the exact task I was asking about. Of course. this time, I placed markers on links to have some atoms to calculate distance with.

When I use the command of

findclash sel test self overlapCutoff -90 hbondAllowance 0 bondSeparation 50

while selecting all new markers, I can visualize...

something like this!

On Sat, Jul 8, 2017 at 4:47 PM, Elaine Meng <meng@cgl.ucsf.edu> wrote:

Hi Jun Ha Song,
Yes, links are just like bonds, and exist just in terms of connecting those endpoints, so you have to fill them in with atoms/markers for findclash to use.

Q1: sorry, there is no way to specify just bonds in the command line, and even if you did, they wouldn’t be the XYZ-points needed by computations like findclash. There is “bondzone” to make various zoning tools use points along bonds, but as far as I know it doesn’t affect distance measurements. You could try it to make sure:
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/bondzone.html>

Q2: there is “distance” to report just the distance between two specified atoms/markers and “measure distance” to find just the closest distance between 2 sets of atoms and/or surface vertices… are either of those useful in your case?
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/distance.html>
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/measure.html#distance>

For command-line specification, you can use model number and/or residue number (which is apparently the marker ID). When I create a marker file (attached below, test.cmm) from the example given in the Volume Tracer manpage,
<http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/volumepathtracer/volumepathtracer.html#markerfiles>
… one marker is residue 1 and the other is residue 2, but they don’t appear to have atom names. So if the marker set is #0, I can use
sel #0:1
sel #0:2
distance #0:1,2

I don’t know how Tom’s script adds the markers along links (whether they’re in the same model as the original marker set, what residue numbers, etc.)

I hope this helps,
Elaine
-----
Elaine C. Meng, Ph.D.
UCSF Chimera(X) team
Department of Pharmaceutical Chemistry
University of California, San Francisco

> On Jul 8, 2017, at 2:41 PM, Jun Ha Song <junhasong@berkeley.edu> wrote:
>
> Hello,
>
> I want to search whether my models are close enough to be exactly apart from each other by the distance I specify.
>
> findClash appear to be tailor-made for the exact task I was hoping to do, but unfortunately, it appear to only search nearby atoms, not the links, even if I select the entire model which has both markers and links incorporated as its component.
>
> The only way that I can make my entire MarkerSet to be competent for findClash was to use Tom's dividelinks.py script (http://plato.cgl.ucsf.edu/trac/chimera/attachment/wiki/Scripts/dividelinks.py) to add n number of marker s along the link of two marker that I selected.
>
> Q1. Can I ever specify MarkerSet Link by atom-spec commands, to use findClash on the inter-model links as well, not only for the markers?
>
> Q2. Is there any other function than findClash that allow me to find any inter-model (never atoms within the same model) atom or link's distance that I specify exist?
>
> Best,
> Jun Ha Song.