July 2012 - Chimera-dev - RBVI Mailing Lists

Re: [chimera-dev] Chimera/Python sparse matrices
by Greg Couch 06 Dec '13

06 Dec '13

On 8/13/2010 12:06 PM, Forbes J. Burkowski wrote: > What are the steps that need to be taken if the Python used in Chimera is > to have the ability to import functions from SciPy or PySparse? > > Thanks for any advice that you can provide. > > Cheers, > Forbes Burkowski If you want to use prebuilt SciPy binaries with chimera, you need one that is compatible with the version of chimera that chimera is using. Chimera 1.4.x uses Python 2.5, and the daily builds use Python 2.7. You would add those packages to chimera by either (1) given the '--pypath dir' command line option, (2) setting your CHIMERAPTH environment variable (instead of PYTHONPATH, PYTHONPATH is ignored so chimera won't be accidentally broken by other installed Pythons), or (3) add the installation directory to sys.path in Python. If you're not using a daily build on Linux, then you should edit CHIMERA/share/chimeraInit.py and comment out or remove the part that sets the RTLD_GLOBAL flag -- it appears to be no longer needed and its presence breaks SciPy. If you're going to compile from source, which you'll need to do if you're using a daily build, I'd recommend trying PySparse first, since it is a smaller package. Python packages that incorporate C (or C++ or Fortran) code are more difficult than prebuilt binaries because (a) you need a compiler, and (b) if the code calls the Python C API, you need a compatible compiler, ideally the same one as what the rest of Chimera was build with. PySparse needs BLAS and LAPACK libraries and optionally wants UMFPACK and/or SuperLU. On on Redhat and Fedora Linux, you can get the first three libraries by installing the blas-devel, lapack-devel, and suitesparse-devel packages. Chances are you'll be able to install PySparse into chimera's python by running "CHIMERA/bin/pythonXXX setup.py install" in the pysparse source tree (where CHIMERA is the path to where chimera was installed). On Linux, you might have to set the LD_LIBRARY_PATH environment variable to CHIMERA/lib. On Mac OS X, the path to the chimera binary is CHIMERA/Contents/Resources/bin/pythonXXX and the environment variable is DYLD_LIBRARY_PATH. Or you could install PySparse into another Python and use the --pypath/CHIMERAPATH/sys.path approach as given above. Hope this helps, Greg

4 3

Re: [chimera-dev] [Chimera-users] Amber Force Field
by Tom Goddard 17 Oct '12

17 Oct '12

Hi Forbes, I think Eric and Conrad may no more about your problem than I do. But I can tell you my best guess why your calculation is abysmally slow. I think the basic trouble is that useRotamer() decides to replace the existing residue rather than simply update its coordinates. In other words it deletes the old side chain atoms and bonds and makes new ones. Then the new ones don't have charges assigned for energy calculation. So the minimize "prep" option adds hydrogens and computes charges. It does this for the entire molecule because it doesn't know how to prep just one atom. So for each residue and each rotamer tried you end up adding hydrogens and recomputing charges for the whole molecule. Turning prep off in the minimize() call will just cause an error because the charges are missing. The "cache" option in the minimize() call also won't work because the molecule has changed. In fact the minimize code is not smart enough to realize the molecule has new atoms and will try to use the cached MMTK copy of the former molecule and this will probably lead to errors (as you describe). So my 30 minute look at the various parts of the Chimera code without any testing suggests the slowness is because hydrogens/charges are recomputed for every rotamer. The new MMTK molecule object is just the one residue you are calculating the energy of. You realize your code does not take account of any interactions of that residue with any other residues. The other residues are completely ignored. That is what exclres means. So you may as well be asking for the energy of a one residue molecule starting in the rotamer configuration with the actual residue backbone atom positions. I bet this would be very much faster if using a rotamer simply changed existing atom coordinates, then you add hydrogens and charges just one time for all the calculations, and you can cache the MMTK molecule object for each single residue energy calculation which will speed up calculating the different rotamer energies for that residue. Eric is the one who would know why the useRotamer() code deletes atoms and adds new ones instead of just changing the coordinates. It would probably be pretty easy to make a new routine to use instead of useRotamer() that simply copied atom coordinates from the rotamer molecules returned by getRotamer() to the original molecule. Or what about just calculating the energy of the rotamer molecules directly without trying to put them into the protein (with all its other residues ignored)? I see when I use the Rotamer dialog and hide the original molecule that the returned rotamers include CA and N backbone atoms but not C, O. Tom > Hi Tom: > > Thanks for your email of May 4. After some experimenting I got the code > to work but the execution times are deplorable! As I indicated earlier I > want to use the Amber energy calculations to compare various confirmations > involving different rotamers. My problem seems to be that I cannot get a > fast execution because I am unable to cache the results of various > computations and I believe the calculations are starting from scratch for > each rotamer combination. Here is a simplified version of the code that I > am using (in this case working with only one rotamer and essentaily > computing its intrinsic energy (I hope)): > > import chimera > import Rotamers > from Rotamers import getRotamers, useRotamer, NoResidueRotamersError > from MMMD import base > > > def evaluateAmberEnergy(res_L, prot): > > fixedAtoms = set() > molecules = list([prot]) > usingResidues = set(res_L) > > exclres = set(sum((m.residues for m in molecules), [])) > exclres.difference_update(usingResidues) > > def run(minimizer): > minimizer.run() > > minimizer = base.Minimizer(molecules, nsteps=0, stepsize=0.02, > cgsteps=0, cgstepsize=0.02, interval = 10, > fixedAtoms=fixedAtoms, nogui=True, > addhyd=True, callback=run, > exclres=exclres, cache=False, prep=True) > > energy = minimizer._mi.universe.energy() > print "energy value: ", energy > return energy > > #========================================================================== > > # Global variable: > prot = chimera.openModels.open("1CRN", type="PDB")[0] > > for rc in prot.residues: > rotCur_L = getRotamers(rc)[1] > for rotC in rotCur_L: > useRotamer(rc, [rotC]) > energyAmber = evaluateAmberEnergy([rc], prot) > print rc.type, rc.id.position, energyAmber > > > > This is producing a lot of text in the Python Shell and when the program > is run with a large protein it will go for hours... > > Any suggestions for speeding this up would be greatly appreciated. I > tried different settings of cache and prep but got Chimera errors such as > missing C++ atom objects. > > Depressing.... > > Best regards, > Forbes > > > > > > On Fri, 4 May 2012, Tom Goddard wrote: > >> Hi Forbes, >> >> If you want to get the minimized energy within your Python script >> then you need to run the minimization using a Python function call >> rather than a Chimera text command. To figure out how to do that I look >> at the Python code that is called by the Chimera text command in your >> Chimera distribution >> >> chimera/share/MMMD/cmdline.py >> >> or also available online >> >> >> http://plato.cgl.ucsf.edu/trac/chimera/browser/trunk/libs/MMMD/cmdline.py >> >> The minimize() Python function is called using the text command >> arguments. I see it basically does >> >> from MMMD import base >> minimizer = base.Minimizer(molecules, nsteps, stepsize, cgsteps, >> cgstepsize, interval, >> fixedAtoms, memorize, nogui, addhyd, lambda m: >> m.run(), exclres, cache, prep) >> >> Looking at base.py in the same directory and tracking back from where >> "Initial energy..." is printed in file MMTKinter.py I see the energy is >> obtained using >> >> minimizer._mi.universe.energy() >> >> I have not tested this out but perhaps it will move you in the direction >> you want to go. >> >> Tom >> >> >>> Hi: >>> >>> Thanks for the response. The code that you suggested seems to work. I am >>> getting the initial energy reported in the Python Shell along with other >>> lines that are reporting on the activities of the command. >>> >>> Since I am embedding this command in a runCommand statement, I would like >>> to get the initial energy value put into a variable within the Python >>> script instead of the Python Shell window. Is there any way to accomplish >>> this? >>> >>> Cheers, >>> Forbes >>> >>> >>> On Fri, 13 Apr 2012, Elaine Meng wrote: >>> >>>> Hi Forbes, >>>> The minimization tool reports the total energy, not any breakdown into components (electrostatic, VDW, strain) or interactions between sets of atoms. However, the "minimize" command has a "fragment" option for ignoring all but the specified residues, so you could at least limit what is considered the total system. That also makes the calculation faster, although single-point energy calculation (0 steps minimization) should already be pretty fast. >>>> Best, >>>> Elaine >>>> >>>> Example (and see also "nogui true"): >>>> >>>> minimize spec :13.a,17.a fragment true cgsteps 0 nsteps 0 >>>> >>>> <http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/minimize.html> >>>> >>>> On Apr 13, 2012, at 10:38 AM, Conrad Huang wrote: >>>> >>>>> Chimera's "Structure Editing -> Minimize Structure" tool (or the "minimize" command) may be used to do this. The tool reports the initial energy of the system in the Reply Log. If you give zero as the number of steps for both steepest descent and conjugate gradient minimization, then Chimera should return immediately. >>>>> >>>>> Conrad >>>>> >>>>> On 4/13/12 7:09 AM, Forbes J. Burkowski wrote: >>>>>> Hi: >>>>>> >>>>>> I have noticed that scripts can start with: >>>>>> >>>>>> import MMTK >>>>>> import MMTK.ForceFields >>>>>> from MMTK.ForceFields import Amber94ForceField >>>>>> >>>>>> We are doing some side chain packing studies that would benefit from the >>>>>> availablity of "easy" force field calculations. Currently, the algorithm >>>>>> works with useRotamers() to set two neighbouring rotamers and this is >>>>>> followed by a simple energy calculation to evaluate the interaction energy >>>>>> between the two rotamers. Right now, the energy calculation is only a bit >>>>>> more sophisticated than a steric collision detector. >>>>>> >>>>>> Question: Can we use some functions in the MMTK.ForceFields module to get >>>>>> an improved energy calculation? >>>>>> >>>>>> Any suggestions would be appreciated. >>>>>> >>>>>> Cheers, >>>>>> Forbes >>>>>> _______________________________________________ >>>>>> Chimera-users mailing list >>>>>> Chimera-users(a)cgl.ucsf.edu >>>>>> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users >>>>> _______________________________________________ >>>>> Chimera-users mailing list >>>>> Chimera-users(a)cgl.ucsf.edu >>>>> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users >>> _______________________________________________ >>> Chimera-users mailing list >>> Chimera-users(a)cgl.ucsf.edu >>> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users >>> >>

5 7

Re: [chimera-dev] [Chimera-users] Motion of an atom
by Eric Pettersen 11 Jul '12

11 Jul '12

Hi Raju, During the execution of a Python script, Chimera does not redraw the graphics screen until explicitly requested or when the script ends. The simplest way to ask for a redraw is to execute the "wait 1" Chimera command, which you can do from Python with: from chimera import runCommand runCommand("wait 1") So if you embed that in your loop you should be able to see your atom move. Frames are redrawn at a rate of 25 per second (though future versions of Chimera may allow faster redraw rates), so you may want to wait more than 1 frame if the atoms moves too fast. Alternatively, there is also a direct Python call to wait(): from Midas import wait wait(1) FYI, there is another mailing list, chimera-dev, where programming-intensive questions like this one are more appropriate, so that the regular user's eyes don't glaze over. :-) --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu On Jul 10, 2012, at 9:07 AM, Raju Purohit wrote: > Hello friends, > > I need some help here. > > I created a basic model having just an atom and set its coordinates as (0,0,0). Now I want to make it move along a certain path defined mathematically (say, along x axis) and I need to see it move. So I used a for loop. But I couldn't see and motion and soon realised that the loop is very quick in executing and thus I could see the atom at its last coordinates. So I tried to add delay of 0.5 seconds for each loop by importing time module and useing time.sleep. But still I couldn't see any motion and the coordinates of the atom is set only at the end of the for loop. I could not see the intermediate path. I am sure you people must have tried this earlier and this problem has a solution but I just couldn't figure it out. > > Can you please help me out? > > Thank you in advance. > > -- > Raju R.N. > Third Year, B.Tech > Mechanical Engineering > National Institute of Technology Karnataka, Surathkal > > _______________________________________________ > Chimera-users mailing list > Chimera-users(a)cgl.ucsf.edu > http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users

2 3

Re: [chimera-dev] [Chimera-users] Programming within Chimera
by Eric Pettersen 11 Jul '12

11 Jul '12

Though one nice thing in Idle is that you can look the attributes/methods of a class (those defined in the C++ layer at least) with the help() function, e.g. "help(chimera.Atom)". I am working on making the output of help() for the most commonly used classes available on the Chimera web site (so that you don't have to start Chimera to get it) and will probably have that done in a week or so. --Eric Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu On Jul 10, 2012, at 9:11 AM, Conrad Huang wrote: > Well, there is Idle :-) The menu item "Tools -> General Controls -> IDLE" will bring up an Idle window. > > I usually use Idle for testing out snippets of code but prefer to edit scripts outside of Chimera. Keeping the "Open File" dialog up by checking the "Keep dialog up after Open" box (above the Close button) helps a lot. The file you last opened remains selected, so you can repeatedly edit your script and hit the Open button to reload it. All a matter of preference... > > Conrad > > On 7/10/12 7:18 AM, Mario Dejung wrote: >> Hi all, >> is there an easy way to test and develop programs within Chimera? I >> always write python scripts and open these in Chimera… Would be much >> easier to have something like idle or similar to develop scripts and >> test them. >> Or is there a better way I did not found? > >> Looking forward to any suggestions. >> Kind regards >> Mario >> _______________________________________________ >> Chimera-users mailing list >> Chimera-users(a)cgl.ucsf.edu >> http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users >> > > _______________________________________________ > Chimera-users mailing list > Chimera-users(a)cgl.ucsf.edu > http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users >

2 1

Patch to center_of_points()
by Nader Morshed 10 Jul '12

10 Jul '12

It seems that the code in the Segger module for calculating the center of a region breaks when the region contain a large number of points. I don't have the error message on hand, but it complains about calling _contour.affine_transform_vertices with a double, rather than a float. It seems that in regions.py the summed coordinate is cast to a float32, then divided by the number of points, which can cause an implicit cast up to float64's when point_count is large! Here's a simple patch to share/Segger/regions.py that re-orders the calls to cast after division, rather than before: --- old/share/Segger/regions.py 2012-03-19 13:45:34.000000000 -0700 +++ new/share/Segger/regions.py 2012-07-09 14:55:26.241372503 -0700 @@ -873,7 +873,7 @@ plists = [r.points() for r in self.childless_regions()] s = numpy.sum([numpy.sum(plist, axis=0) for plist in plists], axis=0) - com = s.astype(numpy.float32) / self.point_count() + com = (s / self.point_count()).astype(numpy.float32) if transform: tf = self.segmentation.point_transform() import _contour -- Nader Morshed <morshed.nader(a)berkeley.edu>

2 1

Motion of an atom
by Raju Purohit 10 Jul '12

10 Jul '12

Hello friends, I need some help here. I created a basic model having just an atom and set its coordinates as (0,0,0). Now I want to make it move along a certain path defined mathematically (say, along x axis) and I need to see it move. So I used a * for* loop. But I couldn't see and motion and soon realised that the loop is very quick in executing and thus I could see the atom at its last coordinates. So I tried to add delay of 0.5 seconds for each loop by importing *time* module and useing *time.sleep*. But still I couldn't see any motion and the coordinates of the atom is set only at the end of the *for* loop. I could not see the intermediate path. I am sure you people must have tried this earlier and this problem has a solution but I just couldn't figure it out. Can you please help me out? Thank you in advance. -- Raju R.N. Third Year, B.Tech Mechanical Engineering National Institute of Technology Karnataka, Surathkal

1 0

Chimera-dev July 2012