Merge branch 'standardize-atomicdistances-to-use-results' of https://github.com/charity-g/mdanalysis into standardize-atomicdistances-to-use-results

Author: charity-g

benchmarks/benchmarks/analysis/hbond_analysis.py

@@ -0,0 +1,31 @@
+"""
+Benchmark for HydrogenBondAnalysis
+"""
+
+try:
+    from MDAnalysis.analysis.hydrogenbonds.hbond_analysis import HydrogenBondAnalysis as HBA
+    from MDAnalysisTests.datafiles import waterPSF , waterDCD
+except ImportError:
+    pass
+
+import MDAnalysis
+
+class HydrogenBondAnalysisBenchmark:
+    """
+    It tests performance of hbond.run() across different
+    number of frames using waterPSF/waterDCD test files.
+    """
+
+    unit = "ms"
+    timeout = 60.0
+    params = [2, 5, 10]
+    param_names = ["n_frames"]
+
+
+    def setup(self, n_frames):
+        u = MDAnalysis.Universe(waterPSF, waterDCD)
+        self.hbonds = HBA(universe= u)
+    
+    def time_run(self, n_frames):
+        self.hbonds.run(stop = n_frames)
+        

package/CHANGELOG

@@ -23,6 +23,7 @@ The rules for this file:
 
 Fixes
  * `MDAnalysis.analysis.atomicdistances.AtomicDistances` results are now consistent with expected `analysis` documentation data type = Results (Issue #4819, PR #5347) Note: This fix is backwards-incompatible.
+ * Fix mixed-case atom types in guess_bonds (Issue #5342, PR #5343)
  * Fixes msd for non-linear frames, when non_linear is not explicitly
    provided (Issue #5100, PR #5254)
  * Fixes TypeError with np.int64 indexing in GSD Reader (Issue #5224)
@@ -46,6 +47,9 @@ Fixes
    DSSP by porting upstream PyDSSP 0.9.1 fix (Issue #4913)
 
 Enhancements
+ * Added documentation for all keyword in select_atoms() and 
+   selections.rst (Issue #5317, PR #5325)
+ * Added HydrogenBondAnalysis benchmark for performance tracking (PR #5309)
  * Added `select=None` in `analysis.rms.RMSD` to perform no selection on
    the input `atomgroup` and `reference` (Issue #5300, PR #5296)
  * MOL2Parser now reads unit cell dimensions from @<TRIPOS>CRYSIN records (Issue #3341)

package/MDAnalysis/core/groups.py

@@ -3289,6 +3289,10 @@ def select_atoms(
 
         **Simple selections**
 
+            all
+                selects all atoms in the current group; the resulting
+                :class:AtomGroup is unique and sorted by index. If the group
+                already corresponds to Universe.atoms, it is returned unchanged.
             protein, backbone, nucleic, nucleicbackbone
                 selects all atoms that belong to a standard set of residues;
                 a protein is identfied by a hard-coded set of residue names so

package/MDAnalysis/guesser/default_guesser.py

@@ -433,6 +433,9 @@ def guess_bonds(self, atoms=None, coords=None):
         else:
             atomtypes = self.guess_types(atom_types=atoms.names)
 
+        # vdwradii keys are uppercase, so normalize atomtypes to match
+        atomtypes = np.char.upper(np.asarray(atomtypes, dtype=str))
+
         # check that all types have a defined vdw
         if not all(val in vdwradii for val in set(atomtypes)):
             raise ValueError(

testsuite/MDAnalysisTests/guesser/test_default_guesser.py

@@ -280,6 +280,19 @@ def test_guess_bonds_peptide():
     assert_equal(np.sort(u.bonds.indices, axis=0), np.sort(bonds, axis=0))
 
 
+def test_guess_bonds_mixed_case_types():
+    """Test that the guesser can handle mixed-case types"""
+    n_atoms = 2
+    # Actual positions don't matter for this test
+    positions = np.array([[0.0, 0.0, 0.0], [2.5, 0.0, 0.0]])
+    # Use mixed-case types
+    topology = Topology(n_atoms, attrs=[Atomtypes(["Au", "h"])])
+    u = mda.Universe(topology)
+    guesser = DefaultGuesser(None)
+    # This should not raise an error due to the mixed-case types
+    bonds = guesser.guess_bonds(u.atoms, positions)
+
+
 @pytest.mark.parametrize(
     "smi",
     [