Mark
On Oct 21, 2008, at 10:53 PM, Jonathan Winger wrote: Hi Mark,I compared the names of the H atoms for each of the residue types in my pdb file with their cif files in the coot monomer library (for example, see below). Then, to make the H names in the pdb file match those in the cif file, I used a text editor to search my pdb file for the string "HB2 SER" and replaced all with "HB1 SER". Next, I replaced all "HB3 SER" with "HB2 SER". (It's important to start by changing the H2s to H1s, then the H3s to H2s. Otherwise, everything ends up as a H1). I did the same sort of thing for residues with other H types, like Arg (with HG2 and 3, HD2 and 3, etc.) After doing that for all residue types, Coot could real-space refine the H atoms without any problem. It is a major PITA, though, especially if you have alternate conformations. Plus, I don't know what it will do as far as compatibility with other software down the line. So far the edited pdb files seem to work ok in phenix, but I'm still checking...
Hope this helps, Jon
From my pdb after adding H with Molprobity/reduce: ATOM 179 N SER A 491 -58.459 56.155 -4.915 1.00101.27 A N ATOM 180 CA SER A 491 -59.811 56.265 -5.435 1.00106.61 A C ATOM 181 C SER A 491 -60.148 57.722 -5.726 1.00108.97 A C ATOM 182 CB SER A 491 -60.812 55.672 -4.442 1.00110.28 A C ATOM 183 OG SER A 491 -62.063 55.428 -5.058 1.00114.41 A O ATOM 184 O SER A 491 -60.682 58.044 -6.786 1.00113.36 A O ATOM 0 H SER A 491 -58.405 55.839 -3.968 1.00101.27 A H new ATOM 0 HA SER A 491 -59.874 55.697 -6.375 1.00106.61 A H new ATOM 0 HB2 SER A 491 -60.413 54.733 -4.032 1.00110.28 A H new ATOM 0 HB3 SER A 491 -60.945 56.362 -3.596 1.00110.28 A H new ATOM 0 HG SER A 491 -62.695 55.040 -4.387 1.00114.41 A H new
From Coot monomer lib ( /sw/share/coot/lib/data/monomers/s/SER.cif): data_comp_SER # loop_ _chem_comp_atom.comp_id _chem_comp_atom.atom_id _chem_comp_atom.type_symbol _chem_comp_atom.type_energy _chem_comp_atom.partial_charge SER N N NH1 -0.204 SER H H HNH1 0.204 SER CA C CH1 0.058 SER HA H HCH1 0.046 SER CB C CH2 -0.040 SER HB1 H HCH2 0.053 SER HB2 H HCH2 0.053 SER OG O OH1 -0.448 SER HG H HOH1 0.302 SER C C C 0.318 SER O O O -0.422
On Oct 21, 2008, at 7:13 PM, Mark Collins wrote: hi Jon What do you mean search and replace? Were the names of some hydrogens incorrect. Feel free to include a few examples. Mark On Oct 21, 2008, at 3:05 PM, Jonathan Winger wrote: Hi Mark and Paul,
I had to manually go through my pdb file and search and replace all the offending hydrogens. Kind of painful. I hope there's some way to either standardize or automate something here, or it's going to be rough interconverting everything all the time...
Best of luck, Jon
On Oct 21, 2008, at 6:59 AM, Paul Emsley wrote:
Hi Mark,
I thought we did - but maybe not :-/
If you could send me an offending residue (just a PDB snippet, no map required) I'll try to see what the issue is.
Regards,
Paul.
p.s. I'll be in Manhattan tomorrow (I am teaching at Cold Spring Harbor today) maybe going to Broadway. You're not a million miles away, right?
Mark Collins wrote: Hi Jonathan Did you find/get a fix for this? I just switched to pdb v3 file, from the happy and well behaved v2 file. It seems that all H_3 position Hydrogens are the problem (ie HB3, HG3 etc.) and as descirbed fly off when using RSR. Thanks Mark (os 10.4 G4, coot 0.5) On Sun, 5 Oct 2008, Jonathan Winger wrote: I'm building a structure in Coot 0.5. Upon real-space refine or regularize, some of the hydrogens on some residues fly off. It doesn't seem to be the old pdb v3 atom-name issue, since other residues in the structure with exactly the same atom names real-space refine without any trouble. Anyone have a fix for this?
Any help would be greatly appreciated, Jon
---------------------------------------------------------------------------------------------- The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' (I found it!) but 'That's funny ...' Isaac Asimov (1920 - 1992) ----------------------------------------------------------------------------------------------
Mark Collins Columbia University Dept. of Biochemistry & Molecular Biophysics Hendrickson Lab, Black Building 259/201 Office/Lab 212 305 1951 (work)
---------------------------------------------------------------------------------------------- The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' (I found it!) but 'That's funny ...' Isaac Asimov (1920 - 1992) ----------------------------------------------------------------------------------------------
Mark Collins Columbia University Dept. of Biochemistry & Molecular Biophysics Hendrickson Lab, Black Building 259/201 Office/Lab 212 305 1951 (work)
|
pdb4coot.sed