Janin analysis does not filter CYSH

[orbeckst]

Expected behavior

The analysis.dihedral.Janin analysis removes known problematic residues. (Or fails with an intelligible error message.)

Actual behavior

When using the test TPR and XTC trajectory, the analysis fails with

ValueError: Too many or too few atoms selected. Check for missing or duplicate atoms in topology.

Code to reproduce the behavior

import MDAnalysis as mda; 
from MDAnalysis.tests import datafiles as data; 
from MDAnalysis.analysis import dihedrals

# GRO works because residue 76 is "CYS"
u = mda.Universe(data.GRO, data.XTC)
J = dihedrals.Janin(u.select_atoms("protein")).run()

# TPR fails because residue 76 is "CYSH"
u2 = mda.Universe(data.TPR, data.XTC)
J2 = dihedrals.Janin(u2.select_atoms("protein")).run()

Current version of MDAnalysis

• Which version are you using? (run python -c "import MDAnalysis as mda; print(mda.__version__)") develop
• Which version of Python (python -V)? 3.7
• Which operating system? macOS