Merge pull request #1476 from haddocking/1475-automated-toppar-generation-for-unknown-ligands

add automated toppar generation for unknown small molecules

Author: rvhonorato

.github/workflows/ci.yml

@@ -17,36 +17,26 @@ on:
     - cron: "0 */6 * * *" # every 6 hours
 
 jobs:
-  ci:
-    runs-on: ${{ matrix.os }}
+  # Full suite on Linux + Python 3.14: unit tests, coverage, integration, e2e, docs.
+  test-linux-py314:
+    runs-on: ubuntu-latest
     permissions:
       contents: read
       actions: read
       checks: write
-    strategy:
-      matrix:
-        os: [ubuntu-latest, macos-latest]
-        python-version: ["3.9", "3.10", "3.11", "3.12", "3.13", "3.14"]
-      fail-fast: false
 
     steps:
       - uses: actions/checkout@v4
 
       - uses: actions/setup-python@v5
         with:
-          python-version: ${{ matrix.python-version }}
+          python-version: "3.14"
 
-      - name: install system dependencies (Linux)
-        if: runner.os == 'Linux'
+      - name: install system dependencies
         run: |
           sudo apt-get update
           sudo apt-get install -y openmpi-bin libopenmpi3 libopenmpi-dev
 
-      - name: install system dependencies (macOS)
-        if: runner.os == 'macOS'
-        run: |
-          brew install open-mpi
-
       - name: install haddock3 with extra dependencies
         run: pip install -e '.[mpi,dev,docs,notebooks]'
 
@@ -55,23 +45,77 @@ jobs:
           pytest -v --random-order tests/
           --cov --cov-report=
 
-      - name: run integration tests
-        run: >-
-          pytest -v --random-order integration_tests/
-
       - name: generate coverage report
         run: |
           coverage report
           coverage xml
 
       - uses: codacy/codacy-coverage-reporter-action@v1
-        if: matrix.python-version == '3.14'
         with:
           project-token: ${{ secrets.CODACY_PROJECT_TOKEN }}
           coverage-reports: ./coverage.xml
 
+      - name: run integration tests
+        run: >-
+          pytest -v --random-order integration_tests/
+
+      - name: run end to end tests
+        run: >-
+          pytest -v --random-order --no-cov end-to-end_tests/
+
       - name: check if docs are buildable
         continue-on-error: true
         run: |
           sphinx-apidoc -f -e -o docs/ src/haddock -d 1
           sphinx-build docs haddock3-docs
+
+  # Run the remaining OSxPython combinations in parallel
+  test-matrix:
+    runs-on: ${{ matrix.os }}
+    permissions:
+      contents: read
+      actions: read
+      checks: write
+    strategy:
+      matrix:
+        # ubuntu-latest    = linux-x86_64
+        # ubuntu-24.04-arm = linux-aarch64
+        # macos-latest     = macos-arm64
+        os: [ubuntu-latest, ubuntu-24.04-arm, macos-latest]
+        # Test only newest and oldest python version
+        # > versions in the middle are unlikely to have compatibility issues
+        python-version: ["3.9", "3.14"]
+        exclude:
+          # Already covered by test-linux-py314, don't run again
+          - os: ubuntu-latest
+            python-version: "3.14"
+      fail-fast: false
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: install system dependencies (Linux)
+        if: runner.os == 'Linux'
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y openmpi-bin libopenmpi3 libopenmpi-dev
+
+      - name: install system dependencies (macOS)
+        if: runner.os == 'macOS'
+        run: |
+          brew install open-mpi
+
+      - name: install haddock3 with extra dependencies
+        run: pip install -e '.[mpi,dev,docs,notebooks]'
+
+      - name: run unit tests
+        run: >-
+          pytest -v --random-order tests/
+
+      - name: run integration tests
+        run: >-
+          pytest -v --random-order integration_tests/

CHANGELOG.md

@@ -1,5 +1,6 @@
 # Changelog
 
+- 2026-02-24: Implement automated toppar generation for unknown atoms with PRODRG
 - 2026-02-20: Add fallback routine to use `Scheduler` if the GRID is not available
 - 2025-12-15: Added missing NGA glycan parameters - Issue #1462
 - 2025-11-25: Simplify the use of multiple ambig archives

MANIFEST.in

@@ -28,4 +28,5 @@ recursive-include varia *.lua
 recursive-include varia *.md
 include src/haddock/bin/*
 include src/haddock/cns/bin/*
+include src/haddock/prodrg/*
 include src/haddock/libs/assets/*

end-to-end_tests/test_protein_ligand.py

@@ -0,0 +1,55 @@
+"""End-to-end test for the protein-ligand docking example."""
+
+import shutil
+import tempfile
+from pathlib import Path
+
+from haddock.clis.cli import main as cli_main
+
+EXAMPLE_DIR = (
+    Path(__file__).resolve().parents[1] / "examples" / "docking-protein-ligand"
+)
+
+
+def test_protein_ligand_autotoppar_workflow(monkeypatch):
+    """Test protein-ligand docking example with automated topology generation.
+
+    Uses docking-protein-ligand-autotoppar-test.cfg to run the full pipeline
+    (topoaa, rigidbody, caprieval, seletop, flexref, caprieval, rmsdmatrix,
+    clustrmsd, seletopclusts, caprieval x2) without any user-provided ligand
+    topology or parameter files.  ``autotoppar=True`` instructs topoaa to
+    invoke prodrg automatically, and the generated files are propagated to all
+    downstream modules via ``_output_params``.
+    """
+    with tempfile.TemporaryDirectory() as tmpdir:
+        shutil.copytree(Path(EXAMPLE_DIR, "data"), Path(tmpdir, "data"))
+        cfg = Path(tmpdir, "docking-protein-ligand-autotoppar-test.cfg")
+        shutil.copy(
+            Path(EXAMPLE_DIR, "docking-protein-ligand-autotoppar-test.cfg"), cfg
+        )
+
+        monkeypatch.chdir(tmpdir)
+        cli_main(cfg)
+
+        run_dir = Path("run1-autotoppar-test")
+
+        # Check if the auto-generated prodrg topology files were generated
+        autotoppar_param = Path(
+            run_dir, "00_topoaa", "oseltamivir_zwitterion_prodrg.param"
+        )
+        autotoppar_top = Path(run_dir, "00_topoaa", "oseltamivir_zwitterion_prodrg.top")
+        assert autotoppar_param.exists(), f"{autotoppar_param} was not generated"
+        assert autotoppar_top.exists(), f"{autotoppar_top} was not generated"
+
+        # Verify all workflow steps produced output directories
+        assert Path(run_dir, "00_topoaa").exists()
+        assert Path(run_dir, "01_rigidbody").exists()
+        assert Path(run_dir, "02_caprieval").exists()
+        assert Path(run_dir, "03_seletop").exists()
+        assert Path(run_dir, "04_flexref").exists()
+        assert Path(run_dir, "05_caprieval").exists()
+        assert Path(run_dir, "06_rmsdmatrix").exists()
+        assert Path(run_dir, "07_clustrmsd").exists()
+        assert Path(run_dir, "08_seletopclusts").exists()
+        assert Path(run_dir, "09_caprieval").exists()
+        assert Path(run_dir, "10_caprieval").exists()

examples/docking-protein-ligand/docking-protein-ligand-autotoppar-test.cfg

@@ -0,0 +1,68 @@
+# ====================================================================
+# Protein-ligand docking example
+
+# directory in which the scoring will be done
+run_dir = "run1-autotoppar-test"
+
+# execution mode
+mode = "local"
+ncores = 40
+
+# molecules to be docked
+molecules =  [
+    "data/neuraminidase-2BAT.pdb",
+    "data/oseltamivir_zwitterion.pdb"
+    ]
+
+# ====================================================================
+# Parameters for each stage are defined below
+# ====================================================================
+[topoaa]
+autohis = true
+delenph = false
+autotoppar = true
+
+[rigidbody]
+tolerance = 20
+ambig_fname = "data/ambig-active-rigidbody.tbl"
+sampling = 20
+w_vdw = 1.0
+
+[caprieval]
+reference_fname = 'data/target.pdb'
+
+[seletop]
+select = 5
+
+[flexref]
+tolerance = 20
+ambig_fname = "data/ambig-passive.tbl"
+mdsteps_rigid = 0
+mdsteps_cool1 = 0
+
+[caprieval]
+reference_fname = 'data/target.pdb'
+
+[rmsdmatrix]
+resdic_A = [151,152,348,276,156,292,277,222,371,246,406,179,178,227,294,224,119,118]
+resdic_B = [500]
+
+[clustrmsd]
+criterion = "maxclust"
+n_clusters = 2
+
+[seletopclusts]
+top_models = 4
+
+[caprieval]
+reference_fname = 'data/target.pdb'
+
+# Running final caprieval with allatoms parameter set to true to also
+#  include the evaluation of protein side chains
+#  in both the alignment process and irmsd, ilrmsd computations
+# NOTE that all ligand atoms are always considered even without this option.
+[caprieval]
+allatoms = true
+reference_fname = "data/target.pdb"
+
+# ====================================================================

integration_tests/test_alascan.py

@@ -8,7 +8,7 @@
 from haddock.modules.analysis.alascan import (
     DEFAULT_CONFIG as DEFAULT_ALASCAN_CONFIG,
     HaddockModule as AlascanModule,
-    )
+)
 from haddock.modules.analysis.alascan.scan import RES_CODES
 from haddock.libs.libio import read_from_yaml
 from haddock.libs.libontology import PDBFile
@@ -40,11 +40,11 @@ def alascan_module_protlig():
         shutil.copy(
             Path(GOLDEN_DATA, "ligand.top"),
             Path(alascan.path, "ligand.top"),
-            )
+        )
         shutil.copy(
             Path(GOLDEN_DATA, "ligand.param"),
             Path(alascan.path, "ligand.param"),
-            )
+        )
         # Set the parameters to point the file
         alascan.params["ligand_param_fname"] = Path(alascan.path, "ligand.param")
         alascan.params["ligand_top_fname"] = Path(alascan.path, "ligand.top")
@@ -73,7 +73,7 @@ def retrieve_models(self, individualize: bool = False):
 
     def output(self):
         return None
-    
+
 
 class MockPreviousIO_single_model:
     def __init__(self, path):
@@ -158,7 +158,7 @@ def test_alascan_single_model(alascan_module, mocker):
     # check single complex csv
     df = pd.read_csv(expected_csv, sep="\t", comment="#")
     assert df.shape == (12, 15), f"{expected_csv} has wrong shape"
-    
+
     # there should be several mutants saved to file
     # for each mutation in df, check that the corresponding file exists
     from haddock.libs.libalign import PROT_SIDE_CHAINS_DICT
@@ -170,7 +170,7 @@ def test_alascan_single_model(alascan_module, mocker):
         mut_file = Path(alascan_module.path, f"2oob-{mut_file_identifier}.pdb")
         assert mut_file.exists(), f"{mut_file} does not exist"
         # now let's open the file and check that the mutation is correct
-        
+
         heavy_atoms = []
         with open(mut_file, "r") as f:
             for ln in f:
@@ -179,9 +179,7 @@ def test_alascan_single_model(alascan_module, mocker):
                     if not atom_name.startswith("H"):
                         heavy_atoms.append(atom_name)
         # heavy_atoms should be = PROT_SIDE_CHAINS_DICT["LYS"] (order may vary)
-        assert set(heavy_atoms) == set(
-            PROT_SIDE_CHAINS_DICT["LYS"]
-        ), (
+        assert set(heavy_atoms) == set(PROT_SIDE_CHAINS_DICT["LYS"]), (
             f"Heavy atoms for {mut_file_identifier} are not correct: {heavy_atoms}"
         )
 
@@ -196,7 +194,9 @@ def test_alascan_mutation_resiudes():
 
 def test_alascan_with_ligand_topar(alascan_module_protlig, mocker):
     """Test the use of alascan in presence of a ligand."""
-    alascan_module_protlig.previous_io = MockPreviousIO_protlig(path=alascan_module_protlig.path)
+    alascan_module_protlig.previous_io = MockPreviousIO_protlig(
+        path=alascan_module_protlig.path
+    )
     alascan_module_protlig.run()
 
     expected_csv = Path(alascan_module_protlig.path, "scan_protlig_complex_1.tsv")
@@ -206,7 +206,9 @@ def test_alascan_with_ligand_topar(alascan_module_protlig, mocker):
     assert expected_clt_csv.exists(), f"{expected_clt_csv} does not exist"
 
     # List mutated files
-    mutated_filepaths = list(Path(alascan_module_protlig.path).glob("protlig_complex_1-*.pdb"))
+    mutated_filepaths = list(
+        Path(alascan_module_protlig.path).glob("protlig_complex_1-*.pdb")
+    )
     assert len(mutated_filepaths) >= 1
 
     # Loop over files