fix inclusion of PRO in secondary structure #5065

orbeckst · 2025-06-12T00:22:19Z

Changes made in this Pull Request:

ported fix from PyDSSP 0.9.1 by @ShintaroMinami to analysis.dssp.DSSP (see also Wrong assignment or prolines? ShintaroMinami/PyDSSP#2)
new kwarg ignore_proline_donor=True for DSSP (the new default changes the behavior and implements the fix, False recovers old behavior); the kwarg also exists in PyDSSP
updated docs
minimal regression tests
updated CHANGELOG

PR Checklist

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!)

Developers Certificate of Origin

I certify that I can submit this code contribution as described in the Developer Certificate of Origin, under the MDAnalysis LICENSE.

📚 Documentation preview 📚: https://mdanalysis--5065.org.readthedocs.build/en/5065/

@ShintaroMinami

- fix #4913 - ported fix from PyDSSP 0.9.1 by @ShintaroMinami to analysis.dssp.DSSP (see also ShintaroMinami/PyDSSP#2) - new kwarg ignore_proline_donor=True for DSSP (the new default changes the behavior and implements the fix, False recovers old behavior); the kwarg also exists in PyDSSP - updated docs - minimal regression tests - updated CHANGELOG

orbeckst

This is a quick draft. I'd be more than happy if someone continued and completed it.

(EDIT: ... and I am also grateful for any comments that would help me to continue working on it.)

orbeckst · 2025-06-12T00:26:07Z

package/MDAnalysis/analysis/dssp/dssp.py

+        self._donor_mask: Optional[np.ndarray] = (
+            ag.residues.resnames != "PRO" if ignore_proline_donor else None
+        )


This may not be correct. The code runs ... but I am not sure if I should be masking corresponding atoms.

orbeckst · 2025-06-12T00:27:40Z

package/MDAnalysis/analysis/dssp/pydssp_numpy.py

+         Mask out any hydrogens that should not be considered (in particular HN
+         in PRO). If ``None`` then all H will be used (behavior up to 2.9.0).


These docs should be more specific and state the shape. I just quickly guessed the shape from https://github.com/ShintaroMinami/PyDSSP/blob/e251a43ff8622fe0a555313b1567edce45e789e8/scripts/pydssp#L30

donor_mask = sequence != 'PRO' if args.ignore_proline_donor else None

orbeckst · 2025-06-12T00:29:28Z

package/MDAnalysis/analysis/dssp/pydssp_numpy.py

+        if donor_mask is not None
+        else np.ones(n_atoms, dtype=float)
+    )
+    donor_mask = np.tile(donor_mask[:, np.newaxis], (1, n_atoms))


Is the donor_mask (one element for each residue) really correct for this tiling????

orbeckst · 2025-06-12T00:31:32Z

package/MDAnalysis/analysis/dssp/pydssp_numpy.py

    hbond_map = (np.sin(hbond_map / margin * np.pi / 2) + 1.0) / 2
-    hbond_map = hbond_map * local_mask
+    hbond_map *= local_mask
+    hbond_map *= donor_mask


Is this correct? The original code uses https://github.com/ShintaroMinami/PyDSSP/blob/e251a43ff8622fe0a555313b1567edce45e789e8/pydssp/pydssp_numpy.py#L72

hbond_map = hbond_map * repeat(donor_mask, 'l1 l2 -> b l1 l2', b=b)

(with einops.repeat()). Note that we create our donor_mask with tile so it may already be the right size and shape.

orbeckst · 2025-06-12T00:32:34Z

testsuite/MDAnalysisTests/analysis/test_dssp.py

 def test_file_guess_hydrogens(pdb_filename, client_DSSP):
+    # run 2.9.0 tests (which include PRO)
+    # ignore_proline_donor=False
+    # TODO: update reference data for ignore_proline_donor=True


We should really have correct reference data. About half of the files do not show a difference between ignore_proline_donor=True and ignore_proline_donor=False.

orbeckst · 2025-06-12T00:33:08Z

testsuite/MDAnalysisTests/analysis/test_dssp.py

-    protein = mda.Universe(TPR, XTC).select_atoms("protein")
-    run = DSSP(protein).run(**client_DSSP, stop=10)


These lines seemed superfluous as the atomgroup approach is tested separately.

codecov · 2025-06-12T00:37:37Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.73%. Comparing base (5c7c480) to head (e9df025).

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #5065      +/-   ##
===========================================
+ Coverage    92.68%   92.73%   +0.05%     
===========================================
  Files          180      180              
  Lines        22452    22456       +4     
  Branches      3186     3186              
===========================================
+ Hits         20809    20824      +15     
- Misses        1169     1176       +7     
+ Partials       474      456      -18

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Copilot

Pull Request Overview

This PR fixes the treatment of proline residues in secondary structure assignment by implementing the DSSP algorithm more correctly. The fix ports changes from PyDSSP 0.9.1 that properly handles proline residues by not considering their HN atoms as hydrogen bond donors.

Adds ignore_proline_donor parameter to DSSP class with default True for correct behavior
Updates the hydrogen bond calculation to mask proline donors when requested
Maintains backward compatibility through the new parameter

Reviewed Changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

File	Description
package/MDAnalysis/analysis/dssp/dssp.py	Adds `ignore_proline_donor` parameter and creates donor mask for proline residues
package/MDAnalysis/analysis/dssp/pydssp_numpy.py	Updates hydrogen bond functions to accept and use donor mask for filtering
testsuite/MDAnalysisTests/analysis/test_dssp.py	Adds regression tests and updates existing tests for the new proline handling
package/CHANGELOG	Documents the fix and new parameter

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

Copilot · 2025-09-04T00:49:37Z

testsuite/MDAnalysisTests/analysis/test_dssp.py

    assert (
        first_frame[:10] != last_frame[:10] == avg_frame[:10] == "-EEEEEE---"
    )


The removal of these duplicate lines that created a new protein AtomGroup and ran DSSP again appears to be cleanup of unused code, but the test logic should be verified to ensure the test still properly validates the trajectory functionality.

Suggested change

assert (

first_frame[:10] != last_frame[:10] == avg_frame[:10] == "-EEEEEE---"

)

assert first_frame[:10] != last_frame[:10]

assert last_frame[:10] == avg_frame[:10] == "-EEEEEE---"

Copilot · 2025-09-04T00:49:38Z

package/MDAnalysis/analysis/dssp/pydssp_numpy.py


 def _get_hydrogen_atom_position(coord: np.ndarray) -> np.ndarray:
-    """Fills in hydrogen atoms positions if they are abscent, under the
+    """Fills in hydrogen atoms positions if they are absent, under the


Fixed spelling error: 'abscent' should be 'absent'.

orbeckst · 2025-10-30T23:05:39Z

@marinegor if you can spare a moment to look at this PR then that would be great; I think you're really the person with best insight into the DSSP code. Feel free to tell me that all these changes are rubbish and we need to redo everything.

marinegor · 2025-11-02T21:11:19Z

@orbeckst I don't think it's all rubbish -- when this change was introduced to pydssp, I also made a similar attempt of introducing it to MDAnalysis.

My main concern is how to combine it with a workaround I introduced when PRO fix wasn't yet available. I have a feeling that in principle we should just get rid of that (I mean roughly this part) and rely purely on pydssp. I'll have a closer look.

orbeckst force-pushed the update-dssp-proline-fix branch from b38e9db to 005100f Compare June 12, 2025 00:23

orbeckst commented Jun 12, 2025

View reviewed changes

Merge branch 'develop' into update-dssp-proline-fix

df31c86

orbeckst mentioned this pull request Jun 26, 2025

Update secondary structure assignment in DSSP after an upstream fix #4913

Open

orbeckst added 2 commits July 14, 2025 13:33

Merge branch 'develop' into update-dssp-proline-fix

87810eb

Merge branch 'develop' into update-dssp-proline-fix

cf1f6aa

orbeckst requested review from RMeli and Copilot September 4, 2025 00:49

Copilot AI reviewed Sep 4, 2025

View reviewed changes

orbeckst requested a review from marinegor October 30, 2025 23:02

Merge branch 'develop' into update-dssp-proline-fix

e9df025

		Mask out any hydrogens that should not be considered (in particular HN
		in PRO). If ``None`` then all H will be used (behavior up to 2.9.0).

		protein = mda.Universe(TPR, XTC).select_atoms("protein")
		run = DSSP(protein).run(**client_DSSP, stop=10)

fix inclusion of PRO in secondary structure #5065

Are you sure you want to change the base?

fix inclusion of PRO in secondary structure #5065

Conversation

orbeckst commented Jun 12, 2025 • edited by github-actions bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Checklist

Developers Certificate of Origin

Uh oh!

orbeckst left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

orbeckst Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

orbeckst Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

orbeckst Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

orbeckst Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

orbeckst Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

orbeckst Jun 12, 2025

Choose a reason for hiding this comment

Uh oh!

codecov bot commented Jun 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Reviewed Changes

Uh oh!

Copilot AI Sep 4, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Sep 4, 2025

Choose a reason for hiding this comment

Uh oh!

orbeckst commented Oct 30, 2025

Uh oh!

marinegor commented Nov 2, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

orbeckst commented Jun 12, 2025 •

edited by github-actions bot

Loading

orbeckst left a comment •

edited

Loading

codecov bot commented Jun 12, 2025 •

edited

Loading