Coordinated Water molecules in PDB crystal structure

User discussions
1

Version 2022.2 with thread-MPI and CUDA

I am sure this has been posted or asked before since coordinated water molecules are pretty huge thing in solvation pockets (think Guo/Rizzo from Stony Brook and Her2 :) but if you go looking for “water” in the forum you can’t find this topic as a needle in a haystack of water questions.

This PI3KA structure indicates in their paper there are " X-ray structure revealed that the pyridine nitrogen atom of 8 is part of a hydrogen bond network involving three water molecules and the side chains of residues Y836, D810, D933 and K802, the latter residue"

So when I run solvate these 3 water molecules in my solv.gro are correctly in their intended positions, i.e. they remained the same coordinates as in my PDB file. However when I run genion since the only way to have these waters in the topol.top is to include them as SOL moleculetype despite in the aminoacid.rtp and my gromacs *.gro file they are under the HOH “residuetype”, but something screwy happens as result from genion when I select Group 15 ( SOL) has 236790 elements. These water molecules get moved to some random coordinates, and the first 9 atom coordinates of my ligand 1LT replace the HOH coordinates. I bet there is something easy I am missing. Now just looking, it basically removed these 3 waters and moved all the coordinate atoms up 9 spaces.

 1332HOH     OW21951  12.223  12.982   4.719
 1332HOH    HW121952  12.288  13.014   4.655
 1332HOH    HW221953  12.188  13.061   4.762
 1333HOH     OW21954  12.264  13.041   5.017
 1333HOH    HW121955  12.207  12.967   4.993
 1333HOH    HW221956  12.226  13.117   4.969
 1334HOH     OW21957  12.068  12.807   4.570
 1334HOH    HW121958  11.991  12.814   4.629
 1334HOH    HW221959  12.135  12.857   4.619
    11LT    C0521960  12.472  12.946   3.951
    11LT    C0121961  12.620  13.077   3.851
    11LT    C0621962  12.784  13.233   3.755
    11LT    O0H21963  12.941  13.598   3.744
    11LT    C0I21964  12.353  12.956   4.247
    11LT    C0J21965  12.316  12.966   4.380
    11LT    C0N21966  12.488  13.110   4.427
    11LT    C0K21967  12.532  13.108   4.294
    11LT    C0O21968  12.566  13.189   4.535

The is result. Here are the ligand coordinates after running genion and selecting SOL group that match EXACTLY the 3 original HOH water molecule coordinates (see first 9 lines above). See C05, C01, C06, O0H, C0I, C0J, C0N, C0K, C0O atom names above.

 10691LT    C0521951  12.223  12.982   4.719
 10691LT    C0121952  12.288  13.014   4.655
 10691LT    C0621953  12.188  13.061   4.762
 10691LT    O0H21954  12.264  13.041   5.017
 10691LT    C0I21955  12.207  12.967   4.993
 10691LT    C0J21956  12.226  13.117   4.969
 10691LT    C0N21957  12.068  12.807   4.570
 10691LT    C0K21958  11.991  12.814   4.629
 10691LT    C0O21959  12.135  12.857   4.619

And then my 3 HOH atoms (9 atoms) get renamed to SOL and are way off the coordinates they need to start in.

34386SOL     OW21951   7.295   2.247   6.967
34386SOL    HW121952   7.368   2.316   6.971
34386SOL    HW221953   7.323   2.172   6.908
34387SOL     OW21954   5.841   3.210   5.876
34387SOL    HW121955   5.745   3.211   5.849
34387SOL    HW221956   5.853   3.263   5.960
34388SOL     OW21957   5.691   2.875   7.312
34388SOL    HW121958   5.614   2.842   7.366
34388SOL    HW221959   5.776   2.845   7.354
2

ok let me try this… [gmx-users] how to differentiate crystal water and explicit waters in gromacs ???

So I get it working with this and grompp recognizes my index with my string [ SOL_&_HOH_&_!H* ] but when I start running NVT the output *.gro files revert the HOH back to SOL which is not something I want to have to keep updating. At least the order in topol.top is correct so it works.