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
ag.unwrap should make the group's fragments whole. It does in version 1.1.1.
Actual behavior
In 2.0.0-dev0 an exception sometimes is raised that the group is not contiguous from bonds.
Code to reproduce the behavior
This was first reported by @hejamu in Discord. He has provided me with a LAMMPS example structure for which this occurs, but I have since been able to reproduce the error with 1hvr.pdb, which is already in the tests (with the CONECT datafile reference).
importMDAnalysisasmdafromMDAnalysis.tests.datafilesimportCONECTu=mda.Universe(CONECT)
u.atoms.unwrap()
## ValueError: AtomGroup was not contiguous from bonds, process failed
Current version of MDAnalysis
2.0.0-dev0
Cause
A first inspection seems to indicate that unwrap is selecting the wrong set of atoms as a fragment to unwrap (in the 1hvr example only parts of the molecule have CONECT records). This is maybe related to the changes introduced in #3005, which touched this aspect of the code.
The text was updated successfully, but these errors were encountered:
Hi! I've run into a similar issue before with my own work. It seems like the .pdb file bond connection is not being read. I had to fix by hardcoding some bonds to maintain pbc. Your issue might be similar to what I had experienced.
Expected behavior
ag.unwrap
should make the group's fragments whole. It does in version 1.1.1.Actual behavior
In 2.0.0-dev0 an exception sometimes is raised that the group is not contiguous from bonds.
Code to reproduce the behavior
This was first reported by @hejamu in Discord. He has provided me with a LAMMPS example structure for which this occurs, but I have since been able to reproduce the error with
1hvr.pdb
, which is already in the tests (with theCONECT
datafile reference).Current version of MDAnalysis
2.0.0-dev0
Cause
A first inspection seems to indicate that unwrap is selecting the wrong set of atoms as a fragment to unwrap (in the
1hvr
example only parts of the molecule haveCONECT
records). This is maybe related to the changes introduced in #3005, which touched this aspect of the code.The text was updated successfully, but these errors were encountered: