You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Here you used the reduce function, this is fine for the case of gaff. However, for the case of openff, there maybe an issue.
E.g.,
In the following example, I have two molecules,
the atoms in molecule 1 is labelled as C1, C2, ...
the atoms in molecule 2 is also labelled as C1, C2, ....
However, C1 in mol1 is different from C1 in mol 2.
If we use the reduce function, it seems that only 1 C1 will be left. Do you know if there is a solution to avoid such merge?
In the following example, I expect to see 17+11 = 24 atom types. But it shows 24 after merge.
Because of the change in the specification of parmed and the addition of a feature called deduplicate (paramed/structure.py), identical label atoms of basically different molecular species are marked with a D (see H11D), essentially avoiding confusion.
The one exception is that if all parameters of the atomic type are identical, adding D is not happen.
Therefore, there is basically no problem using current settings for co-crystal.
If you want to completely separate the atomic types between different molecules, you can do so by adding the following code
for i, molecule in enumerate(self.ff.molecules):
for at in molecule.atoms:
at.atom_type.number = i
Currently the atom_type.number is not used. If this value is different for different molecules, D will always be given if the atom_type does not match and the label is the same.
@shinnosukehattori
If we are dealing with a co-crystal, we used the following codes to store the molecular topology information
https://github.com/MaterSim/OST/blob/332976973ea0ccf818245c1a1f0e7975e391f204/ost/parameters.py#L1917-L1924
Here you used the
reduce function
, this is fine for the case ofgaff
. However, for the case ofopenff
, there maybe an issue.E.g.,
In the following example, I have two molecules,
C1
,C2
, ...C1
,C2
, ....However,
C1
in mol1 is different fromC1
in mol 2.If we use the reduce function, it seems that only 1 C1 will be left. Do you know if there is a solution to avoid such merge?
In the following example, I expect to see 17+11 = 24 atom types. But it shows 24 after merge.
The text was updated successfully, but these errors were encountered: