search

openforcefield / smirnoff99Frosst

A general small molecule force field descended from AMBER99 and parm@Frosst, available in the SMIRNOFF format
Creative Commons Attribution 4.0 International
28 stars 9 forks source link

S-Halogen (not in parm99/parm@Frosst) #33

Closed bannanc closed 7 years ago

bannanc commented 7 years ago

S-F bonds show up in drugbank. No other S-halogen bonds are there, but I figure if we're adding one, we should probably add all of them (similar to the halogen - P or N bonds already listed in the issue tracker)

Here's what I've found in GAFF2:

S-F

f -s2  363.47   1.643                     SOURCE2       1
f -s   344.73   1.660                     SOUECE3       1
f -s4  428.80   1.591                     SOURCE2       4    0.0065
f -s6  400.85   1.612             SOURCE2_SOURCE5      15    0.0133
f -sh  356.72   1.649                     SOURCE3       1
f -ss  373.87   1.634                     SOURCE3       3    0.0156

S-Cl

cl-s   230.41   2.072                     SOURCE1       6
cl-s2  185.62   2.161                     SOURCE2       1
cl-s4  230.41   2.072                     SOURCE1       6
cl-s6  230.41   2.072                     SOURCE1       6
cl-sh  230.41   2.072                     SOURCE1       6
cl-ss  230.41   2.072                     SOURCE1       6
cl-sx  230.41   2.072                     SOURCE1       6
cl-sy  230.41   2.072                     SOURCE1       6

S-Br

br-s   169.20   2.220                     SOUECE3       1
br-s4  128.80   2.341                     SOURCE3       1
br-s6  171.57   2.214                     SOURCE3       3    0.0443
br-sh  173.58   2.209                     SOURCE3       1
br-ss  176.02   2.203                     SOURCE3       3    0.0035

P-I

i -s   115.77   2.430                     SOUECE3       1
i -s4   49.21   2.870                     SOUECE3       1
i -s6   49.21   2.870                     SOURCE3       1
i -sh   88.56   2.560                     SOUECE3       1
i -ss   86.63   2.571                     SOURCE3       3    0.0065

Here's what I suggest we add:

[#16:1]-[#9:2]   375.   1.6
[#16:1]-[#17:2]   230.   2.0
[#16:1]-[#35:2]   170.   2.2
[#16:1]-[#53:2]    75.    2.6
davidlmobley commented 7 years ago

Looks good to me. I have to admit I find the S-CL case in GAFF2 somewhat silly because of the redundancy.

bannanc commented 7 years ago

Added in pull request #43