C++ Delphi versions do not seem to understand the radii.siz (default.siz) file used together with previous versions. The beauty (and shortcoming) of the old file was that it only assigned a single radius per element. This does not seem to work any longer.
[x] added to biskit/data/delphi/radii_delphi_v7.siz
[x] adapt atom and residue names in this file to Amber topology
[x] make this the default radii file of Delphi wrapper
This file is optimized to be used with a different charge set. Not sure though, whether it makes any sense physically to combine these radii with amber charges. Better would be to re-create this file from the Amber topology. Radii are not given in the leap prep library files. Instead, they may be fetched from AmberTools/src/pbsa/sa_driver.F90, at least this is what is claimed here:
http://archive.ambermd.org/201610/0088.html
C++ Delphi versions do not seem to understand the radii.siz (default.siz) file used together with previous versions. The beauty (and shortcoming) of the old file was that it only assigned a single radius per element. This does not seem to work any longer.
As a temporary workaround, I am adding the new "sample radius file" from the Delphi web site: https://honiglab.c2b2.columbia.edu/software/DelPhi/doc/parseres.siz
biskit/data/delphi/radii_delphi_v7.sizDelphiwrapperThis file is optimized to be used with a different charge set. Not sure though, whether it makes any sense physically to combine these radii with amber charges. Better would be to re-create this file from the Amber topology. Radii are not given in the leap prep library files. Instead, they may be fetched from
AmberTools/src/pbsa/sa_driver.F90, at least this is what is claimed here: http://archive.ambermd.org/201610/0088.html