- Chimera-users - RBVI Mailing Lists

Fwd: stereo view
by Francesco Pietra 24 Feb '09

24 Feb '09

It was a bad idea. What Jason initially suggested (save left-eye and right-eye) should work. What I have to manipulate images is only gimp, with which I am not very familiar. But it seems to have the capability of arranging the two saved files so as to get a stereo effect. There is even a (unofficial) plugin to remove unwanted space around. francesco ---------- Forwarded message ---------- From: Francesco Pietra <chiendarret(a)gmail.com> Date: Tue, Feb 24, 2009 at 9:31 AM Subject: Re: [Chimera-users] stereo view To: Elaine Meng <meng(a)cgl.ucsf.edu> Hi Elaine: That makes problems as the referee just wants a stereo view of details of the binding site, which requires magnifying the figure. Is it any way to save a pdb file of the magnified zone alone? (to start from there to build a stereoview). Of course, the normal pdb saving of a magnified figure saves the coordinates for the whole. The only way I can conceive to get the coordinates of the binding site, and nothing else, is manually deleting rows from the pdb file. A tricky route because the chains start and return to the starting point and the binding site is on both strands. thanks francesco On Tue, Feb 24, 2009 at 1:27 AM, Elaine Meng <meng(a)cgl.ucsf.edu> wrote: > I guess you could also save separate left eye and right eye images and put > them farther apart. That might also mess up the stereo effect, though. > Elaine > > On Feb 23, 2009, at 4:26 PM, Elaine Meng wrote: > >> Hi Francesco, >> There is a certain separation needed or else your eyes cannot form the >> stereo image. You could try zooming out so the atoms don't take up the >> whole window, or adjusting parameters in the Camera tool (under Tools -> >> Viewing Controls) but I usually don't mess with those camera parameters >> because you could disrupt the stereo viewing if you change them very much. >> Elaine >> >> >> On Feb 23, 2009, at 3:06 PM, Francesco Pietra wrote: >> >>> Hi Elaine: >>> Is it any direct way to save stereoimages from magnified images? By >>> the normal route, the wall-eye couple of images are saved adjacent to >>> one another, with no space in between them. >>> >>> thanks >>> francesco >>> >>> On Tue, Feb 17, 2009 at 1:43 AM, Elaine Meng <meng(a)cgl.ucsf.edu> wrote: >>>> >>>> Hi Francesco, >>>> All of Jason's suggestions are good (thanks Jason!), and I only wanted >>>> to >>>> add a few things: >>>> >>>> (1) while the window will always be a rectangle, you could resize it so >>>> that >>>> there is a different ratio between height and width. If your long >>>> molecule >>>> is oriented across, you could resize the window to be less tall so that >>>> the >>>> width:height ratio is greater. Don't worry if the window is then >>>> filling >>>> less of your screen -- you can specify as high a resolution as you want >>>> in >>>> the File... Save Image dialog. Of course, with a given window size, you >>>> should scale up your molecule(s) to fill it as much as possible. In >>>> agreement with Jason, I think usually a white background is better than >>>> black for publication purposes. How to change background color is >>>> described >>>> near the top of the "Tips on Preparing Images": >>>> <http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/print.html#tips> >>>> >>>> (2) There is a section on saving stereo images near the bottom of that >>>> "Tips >>>> on Preparing Images" page (by the way, you can use "Help... Search >>>> Documentation" to look for help topics such as "stereo image"). I think >>>> journals usually show the "wall-eye" type of stereo, which can be >>>> achieved >>>> by saving one image in the wall-eye stereo mode, or by saving separate >>>> left >>>> and right views and placing them side by side with the left-eye view on >>>> the >>>> left and the right-eye view on the right. >>>> >>>> (3) just show the surface, adjust clipping and depth-cueing as Jason >>>> suggested, and use your artistic and scientific judgement as to whether >>>> it >>>> improves your image! There are additional suggestions for improving >>>> surface >>>> appearance in the "Tips on Preparing Images" page. >>>> >>>> Best, >>>> Elaine >>>> ----- >>>> Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu >>>> UCSF Computer Graphics Lab and Babbitt Lab >>>> Department of Pharmaceutical Chemistry >>>> University of California, San Francisco >>>> http://www.cgl.ucsf.edu/home/meng/index.html >>>> >>>> >>>> >>>> On Feb 16, 2009, at 2:04 PM, odonnell(a)chem.fsu.edu wrote: >>>> >>>>> Hi Francesco, >>>>> A couple of cents before the professionals respond with (very >>>>> likely) better ideas: >>>>> >>>>> In response to 1) Black is best for contrast but uses a lot of ink and >>>>> money. >>>>> White should be fine. >>>>> >>>>> n response to 2) If you go to TOOLS=>VIEWING CONTROLS=>CAMERA, you can >>>>> select various camera modes. CROSS-EYE stereo will put your chimera >>>>> graphic into cross-eye stereo. You can save an image as it is >>>>> displayed on the screen. But i find saving two images, stereo left eye >>>>> and stereo right eye better. Place these side by side in another >>>>> graphic program, (e.g Adobe illustrator) and voila, stereo viewing. >>>>> With two single images you can switch them around to alternate between >>>>> wall-eye and cross-eye. Also, with two images its easier to make the >>>>> distance between them a certain length, as journals are usually >>>>> stringent about how the stereo images are displayed. >>>>> >>>>> In response to 3)I'd to go TOOLS=>VIEWING CONTROLS=> SIDE VIEW and >>>>> slide the back and front yellow planes. This will create a slab >>>>> through your object removing other structural elements which might >>>>> complicate the image. Also, >>>>> try DEPTH CUEING under TOOLS=>VIEWING CONTROLS=> EFFECTS. tweak the >>>>> "start ratio and yon intensity to further remove emphasis on unwanted >>>>> objects. The Depth effects can make some really neat graphics >>>>> >>>>> Hope this helps. >>>>> >>>>> -Jason >>>>> >>>>> Quoting Francesco Pietra <chiendarret(a)gmail.com>: >>>>> >>>>>> Hi: >>>>>> >>>>>> I would greatly appreciate advice about the following requests >>>>>> (between "".."") by the referees (interaction between two proteins; >>>>>> graphics by chimera): >>>>>> >>>>>> 1) "there is way too much wasted black space in figures". >>>>>> >>>>>> As the complex of proteins is elongated along one axis, I found it >>>>>> unavoidable to have wasted space around. Perhaps using a white >>>>>> background in the hope that white is less offending than black? Or is >>>>>> it a way to tailor the background according to the shape of the >>>>>> object? >>>>>> >>>>>> 2) "a stereoview might be helpful for better understanding". >>>>>> >>>>>> How to fulfill this request with chimera I have no idea. >>>>>> >>>>>> 3) "a molecular surface representation of the docking site would be >>>>>> useful". >>>>>> >>>>>> I read on passing about molecular surfaces on this forum. Is that >>>>>> feasible for the highly complex situation of 3D perspective with so >>>>>> many atoms at different "layers"? >>>> >>>> >> > >

1 0

Letter Reference for Amino Acid Atoms
by Sarah Himes 23 Feb '09

23 Feb '09

Hi, My name is Sarah Himes. I'm a graduate student at the University of Maryland. I'm researching alkaline phosphatase for an advanced course. I'm using the file 1ALK in Chimera. I'm pretty familiar with Chimera but I had a question regarding the letter notation of amino acid atoms when you scroll over them. Here are some examples: His370 (NE2) Ser102 (CB) Ser102 (N) Asp327 (OD1) Asp327 (OD2) Glu322 (OE2) My question is, what do the letters in parentheses mean, what do they correspond to? When these letters show up in Chimera, there are no parentheses. I look forward to hearing from you shortly. Thank you for your time. Sincerely, Sarah Himes shimes(a)umd.edu 240-285-1251

2 1

drawing parallelepiped
by Elaine Meng 23 Feb '09

23 Feb '09

On Feb 23, 2009, at 3:31 AM, Francesco Pietra wrote: > > For mono view figures, I wonder whether Chimera allows enclosing the > model in a parallelepiped. This would confer a sense of depth to a > very complex model (a complex of two proteins, partly in a lipidic > membrane. Hi Francesco, There are a few different possibilities: (a) make a PDB file with "atoms" at the corners and the desired bonds (described in CONECT lines), show as wire or stick. The Crystal Contacts tool uses that trick, but is unlikely to generate what you want in this case. You could try creating it interactively with markers and links using Volume Tracer, but probably just creating a PDB file in a text editor would be easier. (b) make a file in BILD format describing the parallelpiped outlines -- requires similar information as the above: coordinates of points at the corners and which points should be connected <http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/bild.html> both (a) and (b) require you to figure out coordinates for the corners yourself... (c) a devious but easy way if you don't mind a rectangular box (all angles 90 degrees) is to use the "molmap" command to create a density map from the displayed atoms, but hide that map's contour surface and just show the box around it. For example, use the command: molmap @/display 2 <http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/molmap.html> Then in the Volume Viewer dialog that appears, choose "Features... Data display options" and turn on "Show outline box" and also choose "File... Remove Surface." In that dialog you can also specify the color and linewidth of the outline. For conveying depth, you can also try adjusting the clipping planes in the Side View. If you put the back one closer to the atoms, the depth cueing "fog" will be more obvious. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html

2 1

coloring the clipped surface
by Elaine Meng 22 Feb '09

22 Feb '09

On Feb 22, 2009, at 4:48 PM, Harinathachari Bahudhanapati wrote: > Dear Elaine, > Chimera is getting more exciting to work with than > ever. I think you guys are the best in the field. > About per-model clipping... > > When I tried, I was not able to find how to color the clipped plane > based on atom B-factors? I was able to give a color (single color) > only > but not based on atom B-factors or other attributes like SAS, SES, > electropotential. > I am not sure if modeled PDB files still have B-factor details. I saw > an image on the chimera website on the feature highlights (B-factor > coloring) and I wondered if I can get few more details on this > feature. > I am trying to make a few pictures for my paper. > Thank you elaine, > sincerely, > Hari > Dear Hari, Thanks for your kind words! Yes, there is a trick to coloring planar surface caps by B-factor or some other attribute, as in that figure: <http://www.cgl.ucsf.edu/chimera/feature_highlights/2gbp-bfactor.png> First, color the atoms by the attribute (Tools... Depiction... Render by Attribute), which would also color the actual molecular surface but not the planar cap where it is clipped. Then, select the atoms and use Color Zone (also under Tools... Depiction) to color the planar cap to match the nearby atoms. Render by Attribute: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/render/ render.html#render> Color Zone: <http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/colorzone/ colorzone.html> I've been wanting to write a tutorial on how to make that figure, but haven't had the time! I also increased the "mesh subdivision factor" in the Surface Capping dialog to 3. Coloring by electrostatic potential is different than coloring by an atom attribute. To color the planar cap by electrostatic potential, you would use Electrostatic Surface Coloring, but it requires a potential map previously created with some other program such as DelPhi. I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html

1 0

multiple coloring
by Anindito Sen 19 Feb '09

19 Feb '09

Dear All Please share with me as how to put multiple color to several pdb structures docked in a density map. Best Andy Dr. Anindito Sen (Ph.D) Research Associate , Dept. of Biochemistry and Molecular Genetics University of Virginia Box 800733 Charlottesville, VA 22908

2 1

Re: [Chimera-users] defining a putative binding site
by Elaine Meng 18 Feb '09

18 Feb '09

Hi Francesco, If you want detailed information about which atoms contact which, and the distances between them, use the "Find Clashes/Contacts" tool, or its command version "findclash", as discussed in several previous posts: <http://www.cgl.ucsf.edu/pipermail/chimera-users/2008-November/ 003260.html> <http://www.cgl.ucsf.edu/pipermail/chimera-users/2008-October/ 003163.html> <http://www.cgl.ucsf.edu/pipermail/chimera-users/2008-October/ 003183.html> There are links in those posts to the manual pages and a tutorial with an example. Some of those discussions were about finding contacts for a ligand, but the findclash feature is general and can be used on whatever set of atoms/residues you want. Again, you can save results to a file or write them to the Reply Log. Best, Elaine ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html On Feb 18, 2009, at 10:55 AM, Francesco Pietra wrote: > Hi Elaine: > > By doing as indicate in your and my post, only residues (45 residues) > for model 0 (the bigger protein) are listed. Then I also commanded: > > sel #1:1701-1793 & 0:20,30-40,55 z<5 > > i.e., inverting the order of the two models. That gave a list of > residues (18 residues) for only model 1 (the smaller protein). > > Probably all that is useful information. But, is any way (one touch, > as above) to get information about which-to-which? > > I am not too lazy, I am prepared on the basis of the above two > listings, to use the zone command residue per residue, should this be > the way. > > (the numbers for the above residues are just for a split example; > actually I set all residues - even for areas that I knoe are not > involved - in the zone command for both proteins) > > Thanks > francesco >

1 0

defining a putative binding site
by Francesco Pietra 18 Feb '09

18 Feb '09

Hi: I need to define a putative binding site for a smaller protein upon a larger one. The task is creating a table of likely interactions between residues of the larger proteinseparated by less than 5 A from those of the smaller protein. Deriving from molecular dynamics, all residues for both proteins are in the same pdb file, alongside lipid, water, etc. I wonder whether the most appropriate way is to separate in two different pdb files the lines belonging to the two different proteins. Then: sel #0:20,30-40,55 & #1:1701-1793 z<5 where "0" is the larger and "1" is the smaller protein. If OK, can the results be obtained in a table form? Thanks francesco pietra

2 1

Re: [Chimera-users] color changes
by Elaine Meng 17 Feb '09

17 Feb '09

On Feb 17, 2009, at 1:15 PM, Jean-Paul Armache wrote: > hello, > i have one question concerning color manipulation in chimera. > every pdb file one opens gets a specific color (if set so). so, > having a few of them open, one can distinguish > between them color-wise. > now, when one selects the files, it is possible to change the color > via actions->color. this indeed changes all > of the things selected, and if the whole pdb was selected, all the > chains there. when one looks then into the > model panel, the old colors are still there - untouched. this would > be reasonable in the case when only parts > of the pdb were selected and re-colored. however if one selects > whole files and changes the colors - is it > possible to have these colors changed also in the model panel? and > is it possible to revert to the old color > once changed through this action menu? > > best regards, > jean-paul armache Hi Jean-Paul, It is because there is a coloring hierarchy: the Model Panel shows the model color, but you can also set atom colors, whichoverrule and hide the model color. As you found, Actions... Color and the "color" command set the atom colors (otherwise you would not be able to make different parts of the same model different colors). If you already colored the atoms of a model, you can "uncolor" some or all of them to reveal the model color using the command "~color". Another way is to choose "Actions... Color... from editor" and then click the "No Color" button in the Color Editor. (The Color Editor can be found in the menu under Tools... Utilities) You can change the model color with the command "modelcolor" or by clicking on the square color wells in the Model Panel to call up the Color Editor. <http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/coloring.html> <http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/hierarchy.html> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. meng(a)cgl.ucsf.edu UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html

1 0

surface view of large file in chimera
by G.-Q. Hu 17 Feb '09

17 Feb '09

Hello, I am trying to show a large file by surface view in chimera. But I found that the structure lost the detail and the surface was smoothed. I guess there is a sampling problem for large file. Is there any way to view the large file at the original resolution? Thanks. Guiqing

2 1

Volume->Fit in Map question
by Keren Lasker 17 Feb '09

17 Feb '09

hello, Is there a way to fit a density map to a pdb rather than a pdb to a density map ? Or in other words, that in the end of the fitting procedure the map would move to best fit the model rather than the other way around. thank you, Keren Lasker.

2 3