In GFN-FF, the EEQ scheme is also used, but in a slightly modified form. In particular, the A matrix is shifted by the Born matrix to account for solvation contributions. Additionally, since GFN-FF operates on fragments, the charge constraint is adapted accordingly.
To make it possible to substitute GFN-FF’s EEQ implementation with multicharge, it seems that some extensions to the current implementation would be necessary.
I would be very interested in seeing support for this use case. Ideally, such an extension could be introduced in a way that does not break the existing API (e.g., via optional parameters or a separate interface).
If this aligns with the project’s direction, I’d be happy to discuss further or help where possible.
In GFN-FF, the EEQ scheme is also used, but in a slightly modified form. In particular, the A matrix is shifted by the Born matrix to account for solvation contributions. Additionally, since GFN-FF operates on fragments, the charge constraint is adapted accordingly.
To make it possible to substitute GFN-FF’s EEQ implementation with multicharge, it seems that some extensions to the current implementation would be necessary.
I would be very interested in seeing support for this use case. Ideally, such an extension could be introduced in a way that does not break the existing API (e.g., via optional parameters or a separate interface).
If this aligns with the project’s direction, I’d be happy to discuss further or help where possible.