With SE-GSM, it is possible to use the option -DQMAG_MAX to specify the maximal step size. This essentially allows to control how fine the path is. Is there an equivalent option in DE-GSM? I tried using -ADD_NODE_TOL, but it does not seem to allow the same control.
Essentially, the problem is that the input structures might be very close or quite different. Currently, DE-GSM uses num_nodes regardless, which makes needlessly fine strings in some cases, while also producing too coarse strings in other cases. I tried playing with some parameters and did not find a reliable way to avoid this.
I am by no means an expert in this domain, but it seems to me like it would be relatively simple to calculate the "distance" in internal coordinates between the two input structures. The number of nodes to use could then be calculated as round(distance/step_size). Such an option would be of great value for my use of pyGSM.
With SE-GSM, it is possible to use the option
-DQMAG_MAXto specify the maximal step size. This essentially allows to control how fine the path is. Is there an equivalent option in DE-GSM? I tried using-ADD_NODE_TOL, but it does not seem to allow the same control.Essentially, the problem is that the input structures might be very close or quite different. Currently, DE-GSM uses
num_nodesregardless, which makes needlessly fine strings in some cases, while also producing too coarse strings in other cases. I tried playing with some parameters and did not find a reliable way to avoid this.I am by no means an expert in this domain, but it seems to me like it would be relatively simple to calculate the "distance" in internal coordinates between the two input structures. The number of nodes to use could then be calculated as
round(distance/step_size). Such an option would be of great value for my use of pyGSM.