Fixed system hierarchy not being fully parsed, added tests/data/protein_small, updated tests

Author: Bernadette-Mohr

src/nomad_simulation_parsers/parsers/gromacs/mdanalysis_parser.py

@@ -473,6 +473,10 @@ def _create_monomer_group(
                 )
                 monomer_group.setdefault('sub_systems', []).append(residue)
 
+            # Set composition formula as RESTYPE(count)
+            monomer_count = len(restype_resids)
+            monomer_group['composition_formula'] = f'{restype}({monomer_count})'
+
             return monomer_group
 
         def _add_residue_hierarchy(

src/nomad_simulation_parsers/parsers/gromacs/parser.py

@@ -540,16 +540,31 @@ def get_subsystems_from_dict(
         Returns:
             List of subsystem dicts from this level
         """
+        # Debug: log what we're receiving
+        source_keys = list(source.keys()) if isinstance(source, dict) else []
+        has_hierarchy = (
+            hasattr(self, '_subsystems_hierarchy') and self._subsystems_hierarchy
+        )
+        self.logger.info(
+            'get_subsystems_from_dict called: source_keys=%s, has_hierarchy=%s',
+            source_keys,
+            has_hierarchy,
+        )
 
         # Root call: return the full hierarchy
         if hasattr(self, '_subsystems_hierarchy') and self._subsystems_hierarchy:
             # Check if this is the root call by looking at source keys
             if 'n_atoms' in source and 'sub_systems' not in source:
+                self.logger.info(
+                    'Returning full hierarchy (%d items)',
+                    len(self._subsystems_hierarchy),
+                )
                 result = self._subsystems_hierarchy
                 return result
 
         # Nested call: extract from dict's sub_systems key
         subsystems = source.get('sub_systems', []) if isinstance(source, dict) else []
+        self.logger.info('Returning nested subsystems (%d items)', len(subsystems))
 
         return subsystems
 
@@ -723,6 +738,7 @@ def get_sub_systems(
 class GromacsArchiveWriter(MDParser):
     def __init__(self, **kwargs):
         self._simulation_parser = GromacsMetainfoParser()
+        self._simulation_parser.max_nested_level = 10
         self._log_parser = GromacsLogParser(text_parser=GromacsLogTextParser())
         self._mdp_parser = GromacsMDPParser(text_parser=GromacsMDPTextParser())
         self._edr_parser = GromacsEDRParser(edr_parser=GromacsEDRFileParser())

src/nomad_simulation_parsers/parsers/utils/mdanalysisparser.py

@@ -114,7 +114,7 @@ def universe(self) -> MDAUniverse | None:
                     self.mainfile, *self.auxilliary_files, **self.options
                 )
             except Exception as e:
-                self.logger.warning('Error creating MDAnalysis universe.', exc_info=e)
+                self.logger.error('Error creating MDAnalysis universe.', exc_info=e)
                 self.universe_error = e
         return self._file_handler
 
@@ -175,9 +175,11 @@ def parse(self, quantity_key: str = None, **kwargs):
             ('names', lambda: ['CGX'] * self.universe.atoms.n_atoms),
             (
                 'moltypes',
-                lambda: self.get_fragtypes()
-                if hasattr(self.universe.atoms, 'fragments')
-                else None,
+                lambda: (
+                    self.get_fragtypes()
+                    if hasattr(self.universe.atoms, 'fragments')
+                    else None
+                ),
             ),
             ('molnums', lambda: getattr(self.universe.atoms, 'fragindices', None)),
             ('resnames', lambda: self._results['atoms_info'].get('resids')),

src/nomad_simulation_parsers/schema_packages/gromacs.py

@@ -124,13 +124,6 @@ class ModelSystem(model_system.ModelSystem):
     add_mapping_annotation(model_system.ModelSystem.velocities, TPR_KEY, '.velocities')
     add_mapping_annotation(model_system.ModelSystem.positions, TPR_KEY, '.positions')
     add_mapping_annotation(model_system.ModelSystem.bond_list, TPR_KEY, '.bond_list')
-    # sub_systems: recursively extract from nested dicts via function call
-    # Pass '.@' as first argument (current node dict) to function
-    add_mapping_annotation(
-        model_system.ModelSystem.sub_systems,
-        TPR_KEY,
-        ('get_subsystems_from_dict', ['.@']),
-    )
     add_mapping_annotation(model_system.AtomsState.m_def, TPR_KEY, '.labels')
 
 
@@ -141,6 +134,13 @@ class ModelSystem(model_system.ModelSystem):
 
 # Subsystem hierarchy annotations (apply to all ModelSystem instances including
 # subsystems)
+# sub_systems: recursively extract from nested dicts via function call
+# Pass '.@' as first argument (current node dict) to function
+add_mapping_annotation(
+    model_system.ModelSystem.sub_systems,
+    TPR_KEY,
+    ('get_subsystems_from_dict', ['.@']),
+)
 add_mapping_annotation(model_system.ModelSystem.name, TPR_KEY, '.name')
 add_mapping_annotation(
     model_system.ModelSystem.composition_formula, TPR_KEY, '.composition_formula'