Missing parameters in current force field

Most of the missing parameters were found in Gen3 torsiondrive set where molecules were prepared by joining different scaffolds, which might be too artificial and may not be relevant. So here is a list of pubchem molecules matching those failed patterns and some solutions to solve those.

Forcefield with the following modifications:

        <!-- modification: b43 changed from [#8X2:1]-[#8X2:2] to [#8X2:1]-[#8X2,#8X1-1:2] -->
        <!-- modification: b53 changed from [#16X2:1]-[#7:2] to [#16X2,#16X1-1:1]-[#7:2] -->
        <!-- modification: a18 changed from [*:1]-[#7X4,#7X3,#7X2-1:2]-[*:3] to [*:1]~[#7X4,#7X3,#7X2-1:2]~[*:3] -->
        <!-- modification: t51 changed from [*:1]-[#6X4:2]-[#7X3:3]-[*:4] to [*:1]~[#6X4:2]-[#7X3:3]~[*:4] -->
        <!-- modification: t130 changed from [*:1]-[#7X4,#7X3:2]-[#7X4,#7X3:3]-[*:4] to [*:1]~[#7X4,#7X3:2]-[#7X4,#7X3:3]~[*:4] -->
        <!-- modification: t138a as a child parameter to include [#7X2]-[#7X4] chemistry, other general force fields can parameterize this -->
        <!-- modification: new parameters for bridgehead Nitrogen chemistry based on t134 and t138 and tweaking the central Nitrogen to be 7x3 -->
        <!-- modification: t161 changed from "[*:1]~[#7X3:2]-[#15:3]~[*:4]" to "[*:1]~[#7:2]-[#15:3]~[*:4]" to make it more general -->

Representative pubchem structure	Error in enumerating valence parameter	Missing SMARTS and probable solution

Representative pubchem structure	Error in enumerating valence parameter	Missing SMARTS and probable solution
https://pubchem.ncbi.nlm.nih.gov/compound/6433205#section=2D-Structure	`Proper torsion assignment failed for [H:7][C:1]([H:8])([H:9])[N+:2](=[N:3][C:4]([H:10])([H:11])[O:5][H:12])[O-:6] ProperTorsionHandler was not able to find parameters for the following valence terms: - Topology indices (2, 1, 0, 6): names and elements ( N), ( N), ( C), ( H), - Topology indices (2, 1, 0, 7): names and elements ( N), ( N), ( C), ( H), - Topology indices (2, 1, 0, 8): names and elements ( N), ( N), ( C), ( H),`	Specific smarts: `[#7X2:1]=[#7X3+1:2]-[#6X4:3]-[:4]`. One solution can be to make `t51: [:1]-[#6X4:2]-[#7X3:3]-[:4]"` more general by `t51_edit: [:1]~[#6X4:2]-[#7X3:3]~[*:4]"` (opls can parameterize)
https://pubchem.ncbi.nlm.nih.gov/compound/67166915	`Angle assignment failed for [H:34][c:13]1[c:12]([c:11]([c:10]([c:15]([c:14]1[H:35])[H:36])[P:8](=[O:9])([N+:16](=[O:17])[O-:18])[O:7][C:6]([H:30])([H:31])[C:5]([H:28])([H:29])[N+:2]([C:1]([H:19])([H:20])[H:21])([C:3]([H:22])([H:23])[H:24])[C:4]([H:25])([H:26])[H:27])[H:32])[H:33] AngleHandler was not able to find parameters for the following valence terms: - Topology indices (7, 15, 16): names and elements ( P), ( N), ( O),`	Specific smarts:`[#8X1:1]=[#7X3+1:2]-[#15X4:3]`. a18 is `"[:1]-[#7X4,#7X3,#7X2-1:2]-[:3]"` , can make it more general `"[:1]~[#7X4,#7X3,#7X2-1:2]~[:3]"` (even opls doesn’t have)
https://pubchem.ncbi.nlm.nih.gov/compound/118984389#section=2D-Structure		Specific smarts pattern: `[#6:1]~[#7X2:2]-[#7X4+1:3]~[#6:4]`. Making `t138: "[:1]~[#7X2:2]-[#7X3:3]~[:4]"` to include `#7X4` as well in the third position, `"[:1]~[#7X2:2]-[#7X3,#7X4:3]~[:4]"`, can be a general solution, but that would be mixing multiplicities, so a new parameter can be `t138a: "[:1]~[#7X2:2]-[#7X4:3]~[:4]"` (opls can parameterize)
https://pubchem.ncbi.nlm.nih.gov/compound/104806#section=2D-Structure		Specific smarts pattern: `"[#8X2:1]-[#8X1-1:2]"`. The bond parameter `b43: "[#8X2:1]-[#8X2:2]"` can be made more general by making it `"[#8X2:1]-[#8X2,#8X1-1:2]"` (even opls doesn’t have)
Only a simple molecule can be found on pubchem matching the [N]-[S-1] pattern, https://pubchem.ncbi.nlm.nih.gov/compound/19977646 Example from the torsion set is below and the corresponding error is in the right column		Specific smarts pattern is `[#16X1-1:1]-[#7X3:2]`. Can be resolved by modifying `b53: "[#16X2:1]-[#7:2]"` to a more general `"[#16X2,#16X1-1:1]-[#7:2]"`. (even opls doesn’t have)
https://pubchem.ncbi.nlm.nih.gov/compound/135436526#section=IUPAC-Name		Specific smarts pattern is `"[:1]-[#7X3:2]-[#7X3+1:3]-[:4]"`. Can be resolved by making `t130 : "[:1]-[#7X4,#7X3:2]-[#7X4,#7X3:3]-[:4]"` more general by making the single bonds flexible, `"[:1]~[#7X4,#7X3:2]-[#7X4,#7X3:3]~[:4]"`. (opls can parameterize)
Larger molecule with this substructure found in ChemBL30		No general torsion term for `[#6X3]-[#16]` central bond. It’s always either #16{X1-1, X2, X3, X3+1 X4] or [#6X3]=[#16]. Can be a new generic parameter before t107.
Found on CheMBL30 (CHEMBL1982357)		Unusual chemistry.
Found on CheMBL30 (CHEMBL331810)		Unusual chemistry
In conjunction with a nitro salt at CHEMBL3391852		Unusual chemistry
Found on CheMBL30		Unusual chemistry, insecticides etc. `t161` can be made more general.
Representative substructure of a larger molecule (along with a salt) found on CheMBL30		There are some occurrences of [S+] in conjunction with salts. Otherwise unusual chemistry.
Representative substructure of a larger molecule found on CheMBL30		Unusual chemistry
		Unusual chemistry