December 2002 - Chimera-dev - RBVI Mailing Lists

Chimera dumps core on this PDB file
by David E. Konerding 30 Jan '03

30 Jan '03

Chimera on my Red Hat system, somewhat understandably but not acceptably in my opinion, dumps core on this PDB file. In CASP, if you can't define a part of the model you submit it with coordinates set to 0. I wonder if Chimera is trying to divide by zero (although I would except a SIGFPE, not a segmentation fault). I did a remote-display from socrates and it had the problem as well; there, I at least got a popup window saying "_chimera: unknown exception" and was able to see this in the reply log: Error processing trigger "addModel": _chimera error: can't normalize zero length array (see reply log for Python traceback info) Traceback (most recent call last): File "/usr/local/chimera/share/__main__.py", line 56, in ? value = chimeraInit.init(sys.argv) File "/usr/local/chimera/share/chimeraInit.py", line 240, in init prefixableType=1) File "/usr/local/chimera/share/chimera/__init__.py", line 628, in open skipViewAll=skipViewAll) File "/usr/local/chimera/share/chimera/__init__.py", line 509, in add self.triggers.activateTrigger(self.ADDMODEL, models) File "/usr/local/chimera/share/chimera/triggerSet.py", line 198, in activateTrigger self._triggerDict[name].activate(data) File "/usr/local/chimera/share/chimera/triggerSet.py", line 120, in activate ret = handler.invoke(self._name, triggerData) File "/usr/local/chimera/share/chimera/triggerSet.py", line 82, in invoke self._funcData, triggerData) File "/usr/local/chimera/share/chimera/__init__.py", line 837, in checkKsdssp model.computeSecondaryStructure(-0.5, 5, 4) _chimera error: can't normalize zero length array shortly after visiting the window a few more times, chimera segfaults. Dave

2 2

Question: pyhton back log & open gl
by manish 14 Jan '03

14 Jan '03

Visit this UCSF Computer Graphics Laboratory message (to reply or unsubscribe) at: http://www.cgl.ucsf.edu/HyperNews/get/chimeraAlphaTest/10.html Thx very much for availing chimera beta version 1 build 1602. I have absolutely no problems in opening pdb files. I am interested in using extension like dock, molecular dynamics. I hope these extension work freely. Can anyone tell me how to see python back log (when i commit some error I am asked to see the python back log). Regarding, collatory hub, I have a lan connection & I have failed to connect (several times). I have opengl with me but i have not made path for it, I mean autoexec.bat (ms dos) which way i should set the path? could u just type the line. Since my wordstar is not working i am not able to install many programs (autoexec.bat) if anyone can send me 2 lines of installation of babel autoexec.bat I would be thankful to him my config--- celeron 500 mhz windows 98 ------------------------------------------------------------- Visit this UCSF Computer Graphics Laboratory message (to reply or unsubscribe) at: http://www.cgl.ucsf.edu/HyperNews/get/chimeraAlphaTest/10.html

2 1

Chimera and Dock
by Eckhart Guthöhrlein 19 Dec '02

19 Dec '02

Hello, our institution was planning to order a copy of MidasPlus, but now we have been told about chimera. A crucial question is: does chimera also include a version of the ms molecular surface calculation program which is suitable for input file preparation for the program DOCK, as it is true for MidasPlus? I could not find the answer in the chimera information available on the web, but I hope that you can help me. Thank you in advance, Eckhart Guthöhrlein

2 1

Re: DelPhi in Chimera
by Thomas Goddard 13 Dec '02

13 Dec '02

Hi Wade, The problem with 1tcr.phi is that it is not the right type of file. It is a text file that gives electric field at each atom position. You can open it in a text editor and see, but Chimera cannot do anything with that file. The DelPhi potential files read by Chimera (*.phi) are binary files that contain a rectangular grid of electric potential values. It looks like the Honig lab server does not provide those. Tom

1 0

Aqua Chimera status
by Thomas Goddard 13 Dec '02

13 Dec '02

Here's a status report on the port of Chimera to the native Mac OS X windowing system Aqua. I have an Aqua Chimera built that can make it through some hand testing showing molecules with different representations and surfaces and show volume data and use some basic user interfaces like side view and model panel and preferences. It also passes all the tests in the Chimera test suite. More testing will be needed to see if everything is working. Showing side view and 1gcn and switching to sphere causes system crash with Radeon 9000. This makes the side view unusable with the Radeon 9000. This crash also occurs on the X windows Mac Chimera. I've reported it to Apple. It does not occur on Eric's GeForce2 MX. To get Tix working I commented out a number of X windows calls and used the standard unix build. Most of this code was #ifdef'ed out for Windows so it should not effect Chimera operation. The one case that is likely to cause problems is an XSetClipRegion() call I commented out. That could cause drawing to run outside the intended bounds. So the Tix port problem looks less severe than I reported before. There are a bunch of Aqua Tk problems. Mouse button 2 and button 3 are exchanged. They are correct in X windows Mac Chimera. Benchmark dialog scrollbar mapped and unmapped continuously when dialog opened. This can disable your whole Mac windowing environment. Sideview multi-tab window keeps OpenGL sideview display on top even when another tab is selected. Mabye a Togl problem. Iconifying main window does not iconify other windows. Pop-up window showing function call arguments in Idle takes focus. Also atom spec popups and toolbar buttons showing popups take focus so button clicks don't go to the desired place. Usually 2 or 3 button clicks works around this. "Data" check button in volume dialog puts "Dat" and "a" on separate lines. Side view "show low res" option menu shows "Fa..." and "T..." instead of False and True. In both cases there is enough space so there is no need for the abbreviations. Clicking on the Idle window puts its menu at the top of the Mac screen. To access the main Chimera menus, need to click on the main graphics window to get them at the top of the Mac screen. Splash screen does not update to show progress messages. It just shows a blank white window until just before it disappears. Tear off menus are in upper left corner of screen under top of screen menu making it impossible to move them. Chimera icon is a separate window. tkwin.wait_visibility() is used by Chimera in splash screen and tkgui.py code and never returns. I commented those out. Requesting multisampling pixel format for side view produces a pixel format but then fails to create graphics context with a bad pixel format error. Need some special case Mac code to work around that bug. I'll try to package up an Aqua Chimera early next week for Eric and Conrad to try on their machines. (I'm working from home tomorrow). Tom

1 0

Re: DelPhi in Chimera
by Thomas Goddard 11 Dec '02

11 Dec '02

Hi Wade, A small 284 Kb 1tcr.phi should not cause Chimera DelPhi Viewer to request too much memory. I suspect the problem is that the DelPhi Viewer extension is not correctly reading the size of the data file from the header. It thinks the phi file is enormous and requests too much memory. The most probable reason for that is the byte order of the DelPhi file is opposite of what is expected. This happens when the Phi file is written on another computer which represents 4 byte integers using the reverse byte order of the computer you are trying to read the data with. Both the Volume Viewer and DelPhi extensions attempt to identify the correct byte order but I am not sure if that works for the DelPhi Viewer. It would be worth trying to open the 1tcr.phi file with the Chimera File/Open... to see if volume viewer can open it. Another possibility is that Chimera really has used up all your computer's memory. How much memory does your machine have? If you email the 1tcr.phi file I can try looking at it with DelPhi Viewer. Tom

1 0

row interleaved stereo support (http://www.vrex.com/developer)
by Jeff Blaney 10 Dec '02

10 Dec '02

Hi, We recently received a VREX twin LCD projector stereo projection system - very cool! Chimera looks great on it. This is the best stereo projection system I've seen. We also got a very clever glass filter that fits over a LCD monitor and enables stereo viewing through passive circularly polarized glasses. It requires row interleaved stereo support, which I currently only have available with VMD. It works great and I think your group and other users would love it. The VRex filter currently is $400 and works with any display card; they currently only support a 15" 1024x768 LCD monitor. Carl Tung from Vrex expects prices will drop; I've got a prototype. It is literally just a glass screen that fits over the LCD screen - no wires or interface of any kind. Would you be willing to add this type of stereo to Chimera? See http://www.vrex.com/developer for details. Thanks, Jeff

2 1

DelPhi in Chimera
by Thomas Goddard 09 Dec '02

09 Dec '02

Hi Wade, If the problem showing the DelPhi file in Chimera is that the file is large you can try giving Chimera more memory. If you are on unix you can use the limit command in a csh shell to see the current limits and to increase the data size limit: % limit cputime unlimited filesize unlimited datasize 131072 kbytes stacksize 8515 kbytes coredumpsize 0 kbytes memoryuse 4092680 kbytes descriptors 4096 files addressspace 4194304 kbytes % limit datasize 512M % limit cputime unlimited filesize unlimited datasize 524288 kbytes stacksize 8515 kbytes coredumpsize 0 kbytes memoryuse 4092680 kbytes descriptors 4096 files addressspace 4194304 kbytes You then need to restart Chimera in that shell where you increased the memory limit. In linux in a bash shell you can use "ulimit -a" to show the limits, and "ulimit -d 500000" to set the limit to 500,000 kbytes = 500 Mbytes. Tom

1 0

Re: Chimera at NCMI
by Thomas Goddard 09 Dec '02

09 Dec '02

Hi Steve, Unfortunately Chimera allows at most one transparent model, with all others opaque if you want correct rendering. As you observed, the models are rendered one after the other, so the second one always appears on top. It would be possible in the future to make volume viewer surfaces render in one model so that all of the surface could be transparent. But then it would not be possible to move one model relative to the other. Would that be useful to you? How important is multiple model transparency to you? Tom > Date: Sat, 7 Dec 2002 14:47:31 -0600 (CST) > From: Steven Ludtke <stevel(a)blake.3dem.bioch.bcm.tmc.edu> > X-X-Sender: stevel(a)id.3dem.bioch.bcm.tmc.edu > Reply-To: sludtke(a)bcm.tmc.edu > To: Thomas Goddard <goddard(a)cgl.ucsf.EDU> > Subject: Re: Chimera at NCMI > In-Reply-To: <200212030001.gB301Ull676766(a)adenine.cgl.ucsf.edu> > Content-Type: TEXT/PLAIN; charset=US-ASCII > X-Filter-Version: 1.9 (cgl.ucsf.edu) > > Hi again. Sorry to keep pestering you. I just ran into another little > problem, and I wanted to find out if its another bug with the NV driver, > or if it's a problem with the vol viewer. > > I just tried to display 2 volume sets simultaneously, and discovered that > it appears that each set is being rendered individually, rather than as > part of the same model. That is, when the structures intersect, I see all > of one structure or all of the other structure rather than proper > clipping/intersecting. Any ideas ? > > TIA! > > -- > ---------------------------------------------------------------------------- > Steven Ludtke, PhD | Baylor College of Medicine > sludtke(a)bcm.tmc.edu | Co-Director > stevel(a)alumni.caltech.edu | National Center For Macromolecular Imaging > V: (713)798-6989 | Dept of Biochemistry and Mol. Biol. > instant messenger: sludtke42 | * Those who do ARE * > http://ncmi.bcm.tmc.edu/~stevel | The converse also holds >

1 0

DelPhi in Chimera
by Thomas Goddard 09 Dec '02

09 Dec '02

Hi Wade, How big is the DelPhi file you are trying to open with DelPhi Viewer in Chimera? Could you send me the Python traceback you get? In addition to the DelPhi Viewer extension you can use the Volume Viewer extension (Extensions/Volumes/Volume Viewer) to look at a DelPhi electrostatics. If you try to open a *.phi file with the Chimera open file dialog it will use Volume Viewer. Volume Viewer can show electrostatic potential isosurfaces. For electrostatics, you may want to show positive and negative isosurfaces in different colors. Using ctrl-click in the Volume Viewer dialog histogram you can add a second threshold level to make a negative potential surface, and you can give it a different color with the color button. One thing the DelPhi Viewer does that Volume Viewer does not do is color a molecular surface. So we would like to get DelPhi Viewer working for you. Tom

1 0

Chimera-dev December 2002