S-P bonds (not in parm99/parm@Frosst)

[bannanc]

This is the first, of I'm sure will be many issues for parameter types that are simply non-existent in the parm@Frosst universe. We would like to get more complete coverage in the force field first then may follow-up later with QM calculations.

Single sulfur phosphorous bonds are not in the parm@Frosst universe.

Here are the relevant bonds I was able to find in GAFF2

p2-s4  126.27   2.190                     SOUECE3       1
p2-s6  129.28   2.180                     SOUECE3       1
p2-sh  217.02   1.971                     SOURCE3      10    0.2829
p2-ss  219.87   1.966                     SOURCE3      10    0.2739
p3-s4  161.75   2.087                     SOURCE3       8    0.2235
p3-s6  165.79   2.077                     SOURCE3      11    0.1420
p3-sh  144.95   2.132                     SOURCE3       3    0.0078
p3-ss  148.86   2.121                     SOURCE3       3    0.0059
p4-s4  109.64   2.251                     SOUECE3       1
p4-s6  105.24   2.269                     SOUECE3       1
p4-sh  151.04   2.115                     SOURCE3       4    0.0008
p4-ss  155.14   2.104                     SOURCE3       4    0.0044
p5-s4  181.84   2.040                     SOUECE3       1
p5-s6  181.84   2.040                     SOUECE3       1
p5-sh  163.76   2.082                     SOURCE3       3    0.0035
p5-ss  150.09   2.118             SOURCE4_SOURCE5      70    0.0117
pe-sx  133.00   2.168                     SOURCE3       9    0.1743
pe-sy  119.67   2.213                     SOURCE3       6    0.0127
pf-sx  133.00   2.168                     SOURCE3       9
pf-sy  119.67   2.213                     SOURCE3       6
px-sx  111.92   2.242                     SOURCE3       3    0.0119
px-sy  110.14   2.249                     SOURCE3       3    0.0272
py-sx  107.66   2.259                     SOURCE3       7    0.0603
py-sy  128.67   2.182                     SOURCE3       5    0.0047

Relevant phosphorous types

p2 30.97         1.538               Phosphate with two connected atoms
p3 30.97         1.538               Phosphate with three connected atoms, such as PH3
p4 30.97         1.538               Phosphate with three connected atoms, such as O=P(CH3)2
p5 30.97         1.538               Phosphate with four connected atoms, such as O=P(OH)3
pe 30.97         1.538               Inner Sp2 P in conjugated systems
pf 30.97         1.538               Inner Sp2 P in conjugated systems, identical to pe
px 30.97         1.538               Special p4 in conjugated systems
py 30.97         1.538               Special p5 in conjugated systems

Relevant sulfur types

s2 32.06         2.900               S with two connected atom, involved at least one double bond
s4 32.06         2.900               S with three connected atoms
s6 32.06         2.900               S with four connected atoms
sh 32.06         2.900               Sp3 S connected with hydrogen
ss 32.06         2.900               Sp3 S in thio-ester and thio-ether
sx 32.06         2.900               Special s4 in conjugated systems
sy 32.06         2.900               Special s6 in conjugated systems

The *x and *y types seem to be fairly specialized so I suggest we use
k = 150
r = 2.1
for now with the SMIRKS [#15:1]-[#16:2] and the "rough estimate from GAFF2 by CCB on [date]"

[davidlmobley]

I totally agree with this, @bannanc and I agree with your estimates; they seem reasonable given the GAFF values and, since this is to maximize coverage with as few parameters as possible, this looks like a good starting point. More specialization can come later when/if the data drives it.

Please start a document indicating prioritized chemistry to look at this summer. Obviously this should be one such region of chemical space.

[bannanc]

added in pull request #43