Hi Emilia, On 03 Jun 2016, at 18:39, Emilia C. Arturo (Emily) <ec...@drexel.edu> wrote:
> Thank you for the reply, Thomas. I have a few items in reply: > > 1) I have not yet tried to change the refinement settings, but (for > educational purposes, if nothing else) I would like to know what it is that > this particular command that you typed, does: PyMOL> refine(5, "mout", > sculpt_field_mask=(0xFFF & ~0x60)). I haven't found any online documentation > for sculpt_field_mask. "sculpt_field_mask" is a bit mask, every bit enables a term. Look at the link to Sculpt.h that I sent in the previous email. 0x60 is the "or" combination of cSculptVDW and cSculptVDW14, and the tilde (~) inverts all bits, so effectively enabling all terms except VDW and VDW14. > 2) Thank you for showing me how to fix atoms by typing > > PyMOL> flag fix, (selection), set > > before I call the morph command. Now how does one UNfix the atoms in the > selection? Navigating Build>Scultping>Clear memory does not do it. Do I > really need to exit the session to accomplish the reset (which is what worked > for me)? "unfix" atoms: PyMOL> flag fix, (selection), clear > 3) I encountered a warning message about the match command when I type: > > PyMOL> morph mout, sele1, sele2, refinement=0 > > (where sele1 and sele2 both have 4 iron ions each, each labeled 'FE') > > The result is several lines that say 'Match-Warning: unknown residue type > 'FE' (using X).' No Fe ions are used/observed in the interpolation. If, > however, I type: > > PyMOL> morph mout, sele1, sele2, match=in, refinement=0 > > then the Fe ions do appear as part of the interpolation. Why does the 'align' > argument (which is default for 'match') ignore the Fe? This is indeed not obvious. The "align" command uses a BLOSUM62 substitution matrix to score the alignment. Since "FE" is not an amino acid, it gets assigned "X" as the single letter code. With the BLOSUM62 matrix, X matched to X has a score of -1 (penalty). In theory you could modify the file /Applications/MacPyMOL.app/Contents/pymol/data/pymol/matrices/BLOSUM62 and change the X/X score to 1, then PyMOL will align the FE atoms. > Thank you for your time, Thomas. > > Regards, > > Emily. Cheers, Thomas > On Tue, May 31, 2016 at 4:44 PM, Thomas Holder > <thomas.hol...@schrodinger.com> wrote: > Hi Emily, > > PyMOL's morphing performs two steps: (1) RigiMOL, and (2) refinement of > non-backbone atoms by sculpting. The second part can be skipped with > "refinement=0". > > Unfortunately, there is no RigiMOL publication or method documentation. > > The sculpting refinement tries to avoid clashes and maintain the local > sidechain geometry, based on start and end conformation as references. You > are right that sculpting doesn't consider the Ramachandran plot or a real > energy force field. But it does consider clashes (it has a VDW term). Check > "Build > Sculpting > ...", the bottom part of that submenu lists some of the > available terms. The corresponding numerical values can be found here: > https://sourceforge.net/p/pymol/code/HEAD/tree/trunk/pymol/layer2/Sculpt.h > > The "morph" command refinement uses all sculpting terms. You could customize > that, e.g. exclude VDW repulsion, like this: > > PyMOL> morph mout, sele1, sele2, refinement=0 > PyMOL> from epymol.rigimol import refine > PyMOL> refine(5, "mout", sculpt_field_mask=(0xFFF & ~0x60)) > > You can also "fix" atoms, e.g. all non-polymer atoms, like free ions, which > tend to bounce around as a result of the refinement: > > PyMOL> morph mout, sele1, sele2, refinement=0 > PyMOL> from epymol.rigimol import refine > PyMOL> flag fix, (not polymer), set > PyMOL> refine(5, "mout") > > Hope that helps. > > Cheers, > Thomas > > On 30 May 2016, at 12:43, Emilia C. Arturo (Emily) <ec...@drexel.edu> wrote: > > > Hello. > > > > I would like to know more about the type of interpolation that is done when > > morph is called, and what calculations are done at each refinement cycle. > > Is there a literature reference or previous forum discussion to which you > > can point me? > > > > From my sifting through the internet, I settled on the likelihood that > > refinement is done using the "molecular sculpting" algorithm that has been > > employed since at least this discussion took place in 2002: > > http://www.mail-archive.com/pymol-users@lists.sourceforge.net/msg00359.html > > In other words, refinement is not done on the basis of energy differences, > > Ramachandran plot, interatomic clashes, or other problems with geometry > > that could arise at each interpolation step. > > > > Is that correct? If so, is there a way to customize the refinement that's > > done? > > > > I am using MacPyMOL version 1.7.6.2. > > > > Regards, > > > > Emily. -- Thomas Holder PyMOL Principal Developer Schrödinger, Inc. ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e _______________________________________________ PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net) Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net
