Merge branch 'develop' into feat/3341-add-crysin-support

Dreamstick9 · web-flow · commit a4c6bb1e13ed · 2026-01-16T18:36:59.000+05:30
diff --git a/.github/PULL_REQUEST_TEMPLATE.md b/.github/PULL_REQUEST_TEMPLATE.md
@@ -8,13 +8,18 @@ Changes made in this Pull Request:
 <!-- Describe the changes that this PR makes. If applicable, use the following bullet list. --> 
 -
 
+## LLM / AI generated code disclosure
+<!-- Please update this disclosure to reflect if you did or did not use LLMs / AI to generate code -->
+LLMs or other AI-powered tools (beyond simple IDE use cases) were used in this contribution: yes / no
+
 ## PR Checklist
 <!-- Please use the following checklist to ensure the PR is ready to be reviewed/merged. -->
  - [ ] Issue raised/referenced?
  - [ ] Tests updated/added?
  - [ ] Documentation updated/added?
  - [ ] `package/CHANGELOG` file updated?
  - [ ] Is your name in `package/AUTHORS`? (If it is not, add it!)
+ - [ ] LLM/AI disclosure was updated.
 
 ## Developers Certificate of Origin
 <!-- 
diff --git a/.github/workflows/deploy.yaml b/.github/workflows/deploy.yaml
@@ -38,7 +38,6 @@ jobs:
       matrix:
         buildplat:
           - [ubuntu-22.04, manylinux_x86_64, x86_64]
-          - [macos-13, macosx_*, x86_64]
           - [windows-2022, win_amd64, AMD64]
           - [macos-14, macosx_*, arm64]
         python: ["cp311", "cp312", "cp313", "cp314"]
diff --git a/.github/workflows/gh-ci-cron.yaml b/.github/workflows/gh-ci-cron.yaml
@@ -140,7 +140,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [ubuntu-22.04, macos-13]
+        os: [ubuntu-22.04, macos-14]
 
     steps:
     - uses: actions/checkout@v4
diff --git a/.github/workflows/linters.yaml b/.github/workflows/linters.yaml
@@ -93,7 +93,6 @@ jobs:
           micromamba: true
           full-deps: true
           numpy: numpy=1.26.0
-          rdkit: rdkit=2023.09.3
 
       - name: install
         run: |
diff --git a/package/CHANGELOG b/package/CHANGELOG
@@ -15,11 +15,18 @@ The rules for this file:
 
 -------------------------------------------------------------------------------
 ??/??/?? IAlibay, orbeckst, marinegor, tylerjereddy, ljwoods2, marinegor,
-         spyke7, talagayev
+         spyke7, talagayev, tanii1125, BradyAJohnston
 
  * 2.11.0
 
 Fixes
+ * HydrogenBondAnalysis: Fixed `count_by_time()` when using `run(FrameIterator)` that
+   results in `self.start` and `self.end` being None (Issue #5200, PR #5202)
+ * NoJump shows a more informative message and fails  when applied
+   outside of the first frame (Issue #4915, PR #5201)
+ * DSSP now explicitly checks for a minimum of 6 residues and raises a clear
+   error message, unlike the previous behavior where it would fail with an
+   incomprehensible broadcasting error at execution time (Issue #5046, PR #5163)
  * Fixes the verbose=False in EinsteinMSD, and only shows progress bar when
    verbose=True (Issue #5144, PR #5153)
  * Fix incorrect assignment of topology_format to format (and vice versa)
diff --git a/package/MDAnalysis/analysis/dssp/dssp.py b/package/MDAnalysis/analysis/dssp/dssp.py
@@ -210,7 +210,7 @@ class DSSP(AnalysisBase):
     Parameters
     ----------
     atoms : Union[Universe, AtomGroup]
-        input Universe or AtomGroup. In both cases, only protein residues will
+        input Universe or AtomGroup with at least 6 protein residues. In both cases, only protein residues will
         be chosen prior to the analysis via `select_atoms('protein')`.
         Heavy atoms of the protein are then selected by name
         `heavyatom_names`, and hydrogens are selected by name
@@ -239,6 +239,8 @@ class DSSP(AnalysisBase):
 
     Raises
     ------
+    ValueError
+        If fewer than 6 residues are provided in the selection.
     ValueError
         if ``guess_hydrogens`` is True but some non-PRO hydrogens are missing.
 
@@ -357,6 +359,12 @@ def __init__(
                 )
             )
 
+        if len(ag.residues) < 6:
+            raise ValueError(
+                "DSSP requires at least 6 residues for secondary structure analysis, "
+                f"but only {len(ag.residues)} residue(s) were provided in the selection."
+            )
+
     def _prepare(self):
         self.results.dssp_ndarray = []
 
diff --git a/package/MDAnalysis/analysis/hydrogenbonds/hbond_analysis.py b/package/MDAnalysis/analysis/hydrogenbonds/hbond_analysis.py
@@ -911,17 +911,13 @@ def count_by_time(self):
              Can be used along with :attr:`HydrogenBondAnalysis.times` to plot
              the number of hydrogen bonds over time.
         """
+        hbond_frames = self.results.hbonds[:, 0].astype(int)
+        frame_unique, frame_counts = np.unique(hbond_frames, return_counts=True)
+        frame_min, frame_max = self.frames.min(), self.frames.max()
 
-        indices, tmp_counts = np.unique(self.results.hbonds[:, 0], axis=0,
-                                        return_counts=True)
-
-        indices -= self.start
-        indices /= self.step
-
-        counts = np.zeros_like(self.frames)
-        counts[indices.astype(np.intp)] = tmp_counts
-
-        return counts
+        counts = np.zeros(frame_max - frame_min + 1, dtype=int)
+        counts[frame_unique - frame_min] = frame_counts
+        return counts[self.frames - frame_min]
 
     def count_by_type(self):
         """Counts the total number of each unique type of hydrogen bond.
diff --git a/package/MDAnalysis/analysis/msd.py b/package/MDAnalysis/analysis/msd.py
@@ -70,7 +70,34 @@
    
    In MDAnalysis you can use the 
    :class:`~MDAnalysis.transformations.nojump.NoJump`
-   transformation. 
+   transformation to unwrap coordinates on-the-fly.
+    
+   A minimal example:
+
+   .. code-block:: python
+      
+      import MDAnalysis as mda
+      from MDAnalysis.transformations import NoJump
+      
+      u = mda.Universe(TOP, TRAJ)
+      
+      # Apply NoJump transformation to unwrap coordinates
+      u.trajectory.add_transformations(NoJump(u))
+       
+      # Now the trajectory is unwrapped and MSD can be computed normally:
+      from MDAnalysis.analysis.msd import EinsteinMSD
+      MSD = EinsteinMSD(u, select="all", msd_type="xyz")
+      MSD.run()
+
+   This example assumes that the trajectory contains periodic box
+   dimensions. If no periodic boundary information is present, box
+   dimensions must be defined before applying ``NoJump``, which can
+   be accomplished by applying the
+   :class:`~MDAnalysis.transformations.boxdimensions.set_dimensions`
+   transformation *before* the 
+   :class:`~MDAnalysis.transformations.nojump.NoJump` transformation.
+   
+   This replaces the need to preprocess trajectories externally.
    
    In GROMACS, for example, this can be done using `gmx trjconv`_ with the
    ``-pbc nojump`` flag.
@@ -382,14 +409,11 @@ def _parse_msd_type(self):
             "xyz": [0, 1, 2],
         }
 
-        self.msd_type = self.msd_type.lower()
-
         try:
-            self._dim = keys[self.msd_type]
-        except KeyError:
+            self._dim = keys[self.msd_type.lower()]
+        except (AttributeError, KeyError):
             raise ValueError(
-                "invalid msd_type: {} specified, please specify one of xyz, "
-                "xy, xz, yz, x, y, z".format(self.msd_type)
+                f"Invalid msd_type {self.msd_type}, must be a string and one of: xyz, xy, xz, yz, x, y, z"
             )
 
         self.dim_fac = len(self._dim)
diff --git a/package/MDAnalysis/analysis/rms.py b/package/MDAnalysis/analysis/rms.py
@@ -244,7 +244,7 @@ def rmsd(a, b, weights=None, center=False, superposition=False):
     >>> _ = u.trajectory[-1]  # forward to last frame
     >>> B = bb.positions.copy()  # coordinates of last frame
     >>> rmsd(A, B, center=True)
-    6.838544558398293
+    np.float64(6.838544558398293)
 
 
     .. versionchanged:: 0.8.1
diff --git a/package/MDAnalysis/lib/util.py b/package/MDAnalysis/lib/util.py
@@ -202,7 +202,7 @@
 import errno
 import functools
 import gzip
-import importlib
+import importlib.util
 import inspect
 import io
 import itertools
diff --git a/package/MDAnalysis/transformations/nojump.py b/package/MDAnalysis/transformations/nojump.py
@@ -118,6 +118,13 @@ def _transform(self, ts):
         except np.linalg.LinAlgError:
             msg = f"Periodic box dimensions are not invertible at step {ts.frame}"
             raise NoDataError(msg)
+
+        if self.prev is None and ts.frame != 0:
+            raise ValueError(
+                "NoJump transformation must be applied starting from frame 0. "
+                f"Currently at frame {ts.frame}. Please reset trajectory to frame 0 before adding this transformation."
+            )
+
         if ts.frame == 0:
             # We don't need to apply the transformation here. However, we need to
             # ensure we have the 0th frame coordinates in reduced form. We also need to
diff --git a/testsuite/MDAnalysisTests/analysis/test_dssp.py b/testsuite/MDAnalysisTests/analysis/test_dssp.py
@@ -89,3 +89,23 @@ def test_exception_raises_with_atom_index(pdb_filename, client_DSSP):
         match="Residue <Residue SER, 298> contains*",
     ):
         DSSP(u, guess_hydrogens=False).run(**client_DSSP)
+
+
+def test_insufficient_residues_raises_error(client_DSSP):
+    """Test that DSSP raises clear error for insufficient residues."""
+    u = mda.Universe(TPR, XTC)
+
+    protein = u.select_atoms("protein")
+
+    with pytest.raises(ValueError, match="DSSP requires at least 6 residues"):
+        res2 = protein.residues[:2].atoms
+        DSSP(res2)
+
+    with pytest.raises(ValueError, match="DSSP requires at least 6 residues"):
+        res4 = protein.residues[:4].atoms
+        DSSP(res4)
+
+    res6 = protein.residues[:6].atoms
+    dssp = DSSP(res6)
+    result = dssp.run(**client_DSSP, stop=1)
+    assert result.results.dssp.shape[1] == 6
diff --git a/testsuite/MDAnalysisTests/analysis/test_hydrogenbonds_analysis.py b/testsuite/MDAnalysisTests/analysis/test_hydrogenbonds_analysis.py
@@ -760,3 +760,23 @@ def test_hbond_analysis(self, universe, client_HydrogenBondAnalysis):
         assert h.hydrogens_sel == ""
         assert h.acceptors_sel == ""
         assert h.results.hbonds.size == 0
+
+
+class TestHydrogenBondAnalysisFrameIterator:
+    @staticmethod
+    @pytest.fixture(scope="class")
+    def universe():
+        return MDAnalysis.Universe(waterPSF, waterDCD)
+
+    def test_frame_iterator(self, universe):
+        frames = np.array([0, 1, 2, 5, 6, 7, 8])
+        hbonds = HydrogenBondAnalysis(
+            universe=universe,
+            hydrogens_sel="name H1 H2",
+            acceptors_sel="name OH2",
+            update_selections=False,
+        )
+        hbonds.run(frames=frames)
+        assert np.array_equal(
+            hbonds.count_by_time(), np.array([2, 1, 4, 3, 3, 2, 2])
+        )
diff --git a/testsuite/MDAnalysisTests/analysis/test_msd.py b/testsuite/MDAnalysisTests/analysis/test_msd.py
@@ -117,9 +117,11 @@ def test_updating_ag_rejected(self, u):
         with pytest.raises(TypeError, match=errmsg):
             m = MSD(updating_ag, msd_type="xyz", fft=False)
 
-    @pytest.mark.parametrize("msdtype", ["foo", "bar", "yx", "zyx"])
+    @pytest.mark.parametrize(
+        "msdtype", ["foo", "bar", "yx", "zyx", 123, "", " xy "]
+    )
     def test_msdtype_error(self, u, SELECTION, msdtype):
-        errmsg = f"invalid msd_type: {msdtype}"
+        errmsg = f"Invalid msd_type {msdtype}, must be a string and one of: xyz, xy, xz, yz, x, y, z"
         with pytest.raises(ValueError, match=errmsg):
             m = MSD(u, SELECTION, msd_type=msdtype)
 
diff --git a/testsuite/MDAnalysisTests/transformations/test_nojump.py b/testsuite/MDAnalysisTests/transformations/test_nojump.py
@@ -379,3 +379,18 @@ def test_notinvertible(nojump_universe):
         ]
         u.trajectory.add_transformations(*workflow)
         transformed_coordinates = u.trajectory.timeseries()[0]
+
+
+@pytest.mark.parametrize("frame_index", [-1, 5])
+def test_nojump_fails_when_not_at_frame_0(frame_index):
+    """
+    Test that NoJump raises a clear error when applied to a trajectory
+    that is not at frame 0.
+    """
+    u = mda.Universe(data.PSF_TRICLINIC, data.DCD_TRICLINIC)
+    u.trajectory[frame_index]
+
+    with pytest.raises(
+        ValueError, match="must be applied starting from frame 0"
+    ):
+        u.trajectory.add_transformations(NoJump())
