RNApolis 0.4.3__py3-none-any.whl → 0.4.6__py3-none-any.whl

{RNApolis-0.4.3.dist-info → RNApolis-0.4.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: RNApolis
-Version: 0.4.3
+Version: 0.4.6
 Summary: A Python library containing RNA-related bioinformatics functions and classes
 Home-page: https://github.com/tzok/rnapolis-py
 Author: Tomasz Zok

{RNApolis-0.4.3.dist-info → RNApolis-0.4.6.dist-info}/RECORD

@@ -4,14 +4,14 @@ rnapolis/common.py,sha256=PUYF01P2vevhyImhZjGYE0jJlsxWHX6GQmsxI4W7S-E,30255
 rnapolis/metareader.py,sha256=I1-cXc2YNBPwa3zihAnMTjEsAo79tEKzSmWu5yvN1Pk,2071
 rnapolis/molecule_filter.py,sha256=hB6-nXgjmw7FAsQ3bj0cZ2FvuW2I1PXunEfcdwEUB1o,7389
 rnapolis/motif_extractor.py,sha256=duHvpi9Ulcny9K60E6VBpz5RpJZw-KdTB4_Ph0iP478,774
-rnapolis/parser.py,sha256=wCA9rXqt51iLECgeBqOShFpuT8JwanNkHYD5uXYvLzU,13988
+rnapolis/parser.py,sha256=2pQYy0sh8TCpeluMmmSJ7C5dudK_bsfstTWCdpwwpNU,15193
 rnapolis/rfam_folder.py,sha256=SjiiyML_T1__saruFwSMJEoQ7Y55GIU8ktS8ZUn5-fw,11111
-rnapolis/tertiary.py,sha256=kysCb3U93Cpxxsfq5uwFxvcI_biwMiRJgWsNZxrXcaA,20501
+rnapolis/tertiary.py,sha256=SQyiYWA0RJhAK70f88CKZvS4EzGKHQ2RoL1s4MueEDQ,21657
 rnapolis/transformer.py,sha256=V9nOQvdq4-p7yUWo0vQg0CDQMpmyxz9t4TMSRVEKHnw,1817
 rnapolis/util.py,sha256=IdquFO3PV1_KDqodjupzm0Rqvgy0CeSzxGHaGEHYXVU,543
-RNApolis-0.4.3.dist-info/LICENSE,sha256=ZGRu12MzCgbYA-Lt8MyBlmjvPZh7xfiD5u5wBx0enq4,1066
-RNApolis-0.4.3.dist-info/METADATA,sha256=24AO4oRKx840RqJLNOF_SrT4xHNtu26s0Nx3URw0NZI,54322
-RNApolis-0.4.3.dist-info/WHEEL,sha256=HiCZjzuy6Dw0hdX5R3LCFPDmFS4BWl8H-8W39XfmgX4,91
-RNApolis-0.4.3.dist-info/entry_points.txt,sha256=foN2Pn5e-OzEz0fFmNoX6PnFSZFQntOlY8LbognP5F0,308
-RNApolis-0.4.3.dist-info/top_level.txt,sha256=LcO18koxZcWoJ21KDRRRo_tyIbmXL5z61dPitZpy8yc,9
-RNApolis-0.4.3.dist-info/RECORD,,
+RNApolis-0.4.6.dist-info/LICENSE,sha256=ZGRu12MzCgbYA-Lt8MyBlmjvPZh7xfiD5u5wBx0enq4,1066
+RNApolis-0.4.6.dist-info/METADATA,sha256=1Zd34oJFXeLnKDzG7HG-6jSjsAnIimsiWPbTlvDerO4,54322
+RNApolis-0.4.6.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
+RNApolis-0.4.6.dist-info/entry_points.txt,sha256=foN2Pn5e-OzEz0fFmNoX6PnFSZFQntOlY8LbognP5F0,308
+RNApolis-0.4.6.dist-info/top_level.txt,sha256=LcO18koxZcWoJ21KDRRRo_tyIbmXL5z61dPitZpy8yc,9
+RNApolis-0.4.6.dist-info/RECORD,,

{RNApolis-0.4.3.dist-info → RNApolis-0.4.6.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (72.2.0)
+Generator: setuptools (75.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

rnapolis/parser.py

@@ -1,7 +1,10 @@
 import logging
 from typing import IO, Dict, List, Optional, Tuple, Union
 
+import numpy as np
 from mmcif.io.IoAdapterPy import IoAdapterPy
+from scipy.spatial import KDTree
+
 from rnapolis.common import ResidueAuth, ResidueLabel
 from rnapolis.tertiary import BASE_ATOMS, Atom, Residue3D, Structure3D
 
@@ -53,10 +56,10 @@ def parse_cif(
 
     io_adapter = IoAdapterPy()
     data = io_adapter.readFile(cif.name)
-    atoms: List[Atom] = []
+    atoms_to_process: List[Atom] = []
     modified: Dict[Union[ResidueLabel, ResidueAuth], str] = {}
-    sequence_by_entity = {}
-    is_nucleic_acid_by_entity = {}
+    sequence_by_entity: Dict[str, str] = {}
+    is_nucleic_acid_by_entity: Dict[str, bool] = {}
 
     if data:
         atom_site = data[0].getObj("atom_site")
@@ -136,7 +139,7 @@ def parse_cif(
                     else None
                 )
 
-                atoms.append(
+                atoms_to_process.append(
                     Atom(
                         label_entity_id,
                         label,
@@ -216,6 +219,7 @@ def parse_cif(
                 if entity_id and pdbx_seq_one_letter_code_can:
                     sequence_by_entity[entity_id] = pdbx_seq_one_letter_code_can
 
+    atoms = filter_clashing_atoms(atoms_to_process)
     return atoms, modified, sequence_by_entity, is_nucleic_acid_by_entity
 
 
@@ -228,7 +232,7 @@ def parse_pdb(
     Dict[str, bool],
 ]:
     pdb.seek(0)
-    atoms: List[Atom] = []
+    atoms_to_process: List[Atom] = []
     modified: Dict[Union[ResidueLabel, ResidueAuth], str] = {}
     model = 1
 
@@ -236,9 +240,6 @@ def parse_pdb(
         if line.startswith("MODEL"):
             model = int(line[10:14].strip())
         elif line.startswith("ATOM") or line.startswith("HETATM"):
-            alternate_location = line[16]
-            if alternate_location != " ":
-                continue
             atom_name = line[12:16].strip()
             residue_name = line[17:20].strip()
             chain_identifier = line[21]
@@ -251,7 +252,10 @@ def parse_pdb(
             auth = ResidueAuth(
                 chain_identifier, residue_number, insertion_code, residue_name
             )
-            atoms.append(Atom(None, None, auth, model, atom_name, x, y, z, occupancy))
+
+            atoms_to_process.append(
+                Atom(None, None, auth, model, atom_name, x, y, z, occupancy)
+            )
         elif line.startswith("MODRES"):
             original_name = line[12:15]
             chain_identifier = line[16]
@@ -263,6 +267,7 @@ def parse_pdb(
             )
             modified[auth] = standard_residue_name
 
+    atoms = filter_clashing_atoms(atoms_to_process)
     return atoms, modified, {}, {}
 
 
@@ -392,3 +397,36 @@ def try_parse_int(s: str) -> Optional[int]:
         return int(s)
     except ValueError:
         return None
+
+
+def filter_clashing_atoms(atoms: List[Atom], clash_distance: float = 0.5) -> List[Atom]:
+    # First, remove duplicate atoms
+    unique_atoms = {}
+
+    for i, atom in enumerate(atoms):
+        key = (atom.label, atom.auth, atom.name)
+        if key not in unique_atoms or atom.occupancy > unique_atoms[key].occupancy:
+            unique_atoms[key] = atom
+
+    unique_atoms_list = list(unique_atoms.values())
+
+    # Now handle clashing atoms
+    coords = np.array([(atom.x, atom.y, atom.z) for atom in unique_atoms_list])
+    tree = KDTree(coords)
+
+    pairs = tree.query_pairs(r=clash_distance)
+
+    atoms_to_keep = set(range(len(unique_atoms_list)))
+
+    for i, j in pairs:
+        if (
+            unique_atoms_list[i].occupancy is None
+            or unique_atoms_list[j].occupancy is None
+        ):
+            continue
+        if unique_atoms_list[i].occupancy > unique_atoms_list[j].occupancy:
+            atoms_to_keep.discard(j)
+        else:
+            atoms_to_keep.discard(i)
+
+    return [unique_atoms_list[i] for i in atoms_to_keep]

rnapolis/tertiary.py

@@ -7,6 +7,7 @@ from typing import Dict, List, Optional, Set, Tuple, Union
 
 import numpy
 import numpy.typing
+
 from rnapolis.common import (
     BasePair,
     BpSeq,
@@ -16,6 +17,7 @@ from rnapolis.common import (
     Residue,
     ResidueAuth,
     ResidueLabel,
+    Saenger,
     Stacking,
 )
 
@@ -521,13 +523,47 @@ class Mapping2D3D:
 
     @cached_property
     def bpseq(self) -> BpSeq:
-        return self.__generate_bpseq(
-            [
-                base_pair
-                for base_pair in self.base_pairs
-                if base_pair.is_canonical and base_pair.nt1 < base_pair.nt2
-            ]
-        )
+        def pair_scoring_function(pair: BasePair3D) -> int:
+            if pair.saenger is not None:
+                if pair.saenger in (Saenger.XIX, Saenger.XX):
+                    return 0, pair.nt1, pair.nt2
+                else:
+                    return 1, pair.nt1, pair.nt2
+
+            sequence = "".join(
+                sorted(
+                    [
+                        pair.nt1_3d.one_letter_name.upper(),
+                        pair.nt2_3d.one_letter_name.upper(),
+                    ]
+                )
+            )
+            if sequence in ("AU", "AT", "CG"):
+                return 0, pair.nt1, pair.nt2
+            return 1, pair.nt1, pair.nt2
+
+        canonical = [
+            base_pair
+            for base_pair in self.base_pairs
+            if base_pair.is_canonical and base_pair.nt1 < base_pair.nt2
+        ]
+
+        while True:
+            matches = defaultdict(set)
+
+            for base_pair in canonical:
+                matches[base_pair.nt1_3d].add(base_pair)
+                matches[base_pair.nt2_3d].add(base_pair)
+
+            for pairs in matches.values():
+                if len(pairs) > 1:
+                    pairs = sorted(pairs, key=pair_scoring_function)
+                    canonical.remove(pairs[-1])
+                    break
+            else:
+                break
+
+        return self.__generate_bpseq(canonical)
 
     def __generate_bpseq(self, base_pairs):
         result: Dict[int, List] = {}