grimme-lab
Reuse of GFN-FF topology in ONIOM calculations
Is it possible to reuse the GFN-FF topology between ONIOM calculations on a large system, where GFN-FF is used as low-precision method? I have tried to run the same calculation twice in the same directory, but it does not look like the program attempts to read an existing topology.
Hi, thank you for your interest in xTB, and sorry for the late reply. While in principle I agree that this would be a nice-to-have I unfortunately do not see a way to implement this cleanly. First, you would have to write two differently named topology files e.g., topo_whole and topo_inner, which would then be read in the next time. However, changing the file names of the topology files is somewhat buried inside the GFN-FF code, and this would mean a lot of code, i.e,. the setup of the FF, would need to be copied and only one line changed. Thus, looking at the current state of the GFN-FF implementation, I do not see a way to do this cleanly.
