Hi Nathan, I have some insights for this issue. Basically the long gaps between the selected parts become too expensive in the alignment scoring. Changing e.g. the gap extension penalty leads to the expected result:
# default is extend=-0.7, lowering to -0.6 super gold & contact_all, test, cycles=0, object=super, extend=-0.6 I think PyMOL's handling of selection gaps during the alignment could be improved, they should be handled different from regular alignment gaps. Cheers, Thomas > On Jun 28, 2018, at 2:14 PM, Thomas Holder <thomas.hol...@schrodinger.com> > wrote: > > Hi Nathan, > > Thank you very much for the files and instructions. I can reproduce the > problem. I haven't seen this before and don't know yet what's going wrong, we > will investigate. > > Cheers, > Thomas > >> On Jun 28, 2018, at 10:50 AM, nclem...@cs.utexas.edu wrote: >> >> The alignment in this example is nearly perfect. Even the residues left >> out of the fitting have perfect matches at the sequence level (as can be >> seen by saving the super object and turning on the sequence alignment and >> contact_all). >> >> PyMOL>load 1AY7_u.pdb, test >> CmdLoad: "1AY7_u.pdb" loaded as "test". >> PyMOL>load 1AY7_b.pdb, gold >> CmdLoad: "1AY7_b.pdb" loaded as "gold". >> PyMOL>select contact_rec, gold & chain R & (all within 10 of gold & chain >> L) & n. ca >> Selector: selection "contact_rec" defined with 33 atoms. >> PyMOL>select contact_lig, gold & chain L & (all within 10 of gold & chain >> R) & n. ca >> Selector: selection "contact_lig" defined with 33 atoms. >> PyMOL>select contact_all, contact_rec + contact_lig >> Selector: selection "contact_all" defined with 66 atoms. >> PyMOL>super gold & contact_all, test, cycles=0 >> MatchAlign: aligning residues (66 vs 185)... >> MatchAlign: score 143.882 >> ExecutiveAlign: 58 atoms aligned. >> Executive: RMS = 0.524 (58 to 58 atoms) >> >> <1AY7_u.pdb><1AY7_b.pdb> -- Thomas Holder PyMOL Principal Developer Schrödinger, Inc. ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ 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
