RNApolis 0.4.4__py3-none-any.whl → 0.4.7__py3-none-any.whl

{RNApolis-0.4.4.dist-info → RNApolis-0.4.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: RNApolis
-Version: 0.4.4
+Version: 0.4.7
 Summary: A Python library containing RNA-related bioinformatics functions and classes
 Home-page: https://github.com/tzok/rnapolis-py
 Author: Tomasz Zok

{RNApolis-0.4.4.dist-info → RNApolis-0.4.7.dist-info}/RECORD

@@ -1,17 +1,17 @@
 rnapolis/annotator.py,sha256=7U3f0gchKdIGc6FwJx0UAc_95HJI5SgECj-b7-1yBhc,22086
 rnapolis/clashfinder.py,sha256=i95kp0o6OWNqmJDBr-PbsZd7RY2iJtBDr7QqolJSuAQ,8513
-rnapolis/common.py,sha256=PUYF01P2vevhyImhZjGYE0jJlsxWHX6GQmsxI4W7S-E,30255
+rnapolis/common.py,sha256=NWhlPwT521jCSWcDcm_TNoYENjoZWpllf9sS-WuTEmA,30361
 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=SQyiYWA0RJhAK70f88CKZvS4EzGKHQ2RoL1s4MueEDQ,21657
+rnapolis/tertiary.py,sha256=6t9ZB4w33-5n_M3sns1RoFXCOTgVAgGH4WDNG5OG9Kg,23426
 rnapolis/transformer.py,sha256=V9nOQvdq4-p7yUWo0vQg0CDQMpmyxz9t4TMSRVEKHnw,1817
 rnapolis/util.py,sha256=IdquFO3PV1_KDqodjupzm0Rqvgy0CeSzxGHaGEHYXVU,543
-RNApolis-0.4.4.dist-info/LICENSE,sha256=ZGRu12MzCgbYA-Lt8MyBlmjvPZh7xfiD5u5wBx0enq4,1066
-RNApolis-0.4.4.dist-info/METADATA,sha256=irtWJbeg1LWun2r3WtnsnDDSHlLvru0hO9wz1e67cIE,54322
-RNApolis-0.4.4.dist-info/WHEEL,sha256=cVxcB9AmuTcXqmwrtPhNK88dr7IR_b6qagTj0UvIEbY,91
-RNApolis-0.4.4.dist-info/entry_points.txt,sha256=foN2Pn5e-OzEz0fFmNoX6PnFSZFQntOlY8LbognP5F0,308
-RNApolis-0.4.4.dist-info/top_level.txt,sha256=LcO18koxZcWoJ21KDRRRo_tyIbmXL5z61dPitZpy8yc,9
-RNApolis-0.4.4.dist-info/RECORD,,
+RNApolis-0.4.7.dist-info/LICENSE,sha256=ZGRu12MzCgbYA-Lt8MyBlmjvPZh7xfiD5u5wBx0enq4,1066
+RNApolis-0.4.7.dist-info/METADATA,sha256=551L8oU_7CdBw7v0jezfHQX7YzF9Fo83E6NVbLVfA50,54322
+RNApolis-0.4.7.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
+RNApolis-0.4.7.dist-info/entry_points.txt,sha256=foN2Pn5e-OzEz0fFmNoX6PnFSZFQntOlY8LbognP5F0,308
+RNApolis-0.4.7.dist-info/top_level.txt,sha256=LcO18koxZcWoJ21KDRRRo_tyIbmXL5z61dPitZpy8yc,9
+RNApolis-0.4.7.dist-info/RECORD,,

{RNApolis-0.4.4.dist-info → RNApolis-0.4.7.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (74.1.2)
+Generator: setuptools (75.6.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

rnapolis/common.py

@@ -338,6 +338,9 @@ class Entry(Sequence):
             return self.pair
         raise IndexError()
 
+    def __lt__(self, other):
+        return self.index_ < other.index_
+
     def __len__(self) -> int:
         return 3
 
@@ -838,7 +841,7 @@ class BpSeq:
 
         for i in range(1, len(regions)):
             k, l, _ = regions[i]
-            available = [True for i in range(10)]
+            available = [True for _ in range(len("([{<" + string.ascii_uppercase))]
 
             for j in range(i):
                 m, n, _ = regions[j]

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

@@ -124,36 +124,17 @@ class Residue3D(Residue):
     outermost_atoms = {"A": "N9", "G": "N9", "C": "N1", "U": "N1", "T": "N1"}
     # Dist representing expected name of atom closest to the tetrad center
     innermost_atoms = {"A": "N6", "G": "O6", "C": "N4", "U": "O4", "T": "O4"}
+    # Heavy atoms in phosphate and ribose
+    phosphate_atoms = {"P", "OP1", "OP2", "O3'", "O5'"}
+    sugar_atoms = {"C1'", "C2'", "C3'", "C4'", "C5'", "O4'"}
     # Heavy atoms for each main nucleobase
     nucleobase_heavy_atoms = {
         "A": set(["N1", "C2", "N3", "C4", "C5", "C6", "N6", "N7", "C8", "N9"]),
         "G": set(["N1", "C2", "N2", "N3", "C4", "C5", "C6", "O6", "N7", "C8", "N9"]),
         "C": set(["N1", "C2", "O2", "N3", "C4", "N4", "C5", "C6"]),
         "U": set(["N1", "C2", "O2", "N3", "C4", "O4", "C5", "C6"]),
+        "T": set(["N1", "C2", "O2", "N3", "C4", "O4", "C5", "C5M", "C6"]),
     }
-    # Heavy atoms in nucleotide
-    nucleotide_heavy_atoms = (
-        set(
-            [
-                "P",
-                "OP1",
-                "OP2",
-                "O5'",
-                "C5'",
-                "C4'",
-                "O4'",
-                "C3'",
-                "O3'",
-                "C2'",
-                "O2'",
-                "C1'",
-            ]
-        )
-        .union(nucleobase_heavy_atoms["A"])
-        .union(nucleobase_heavy_atoms["G"])
-        .union(nucleobase_heavy_atoms["C"])
-        .union(nucleobase_heavy_atoms["U"])
-    )
 
     def __lt__(self, other):
         return (self.model, self.chain, self.number, self.icode or " ") < (
@@ -202,9 +183,59 @@ class Residue3D(Residue):
 
     @cached_property
     def is_nucleotide(self) -> bool:
-        return self.nucleotide_heavy_atoms.intersection(
-            set([atom.name for atom in self.atoms])
+        scores = {"phosphate": 0.0, "sugar": 0.0, "base": 0.0, "connections": 0.0}
+        weights = {"phosphate": 0.25, "sugar": 0.25, "base": 0.25, "connections": 0.25}
+
+        residue_atoms = {atom.name for atom in self.atoms}
+
+        phosphate_match = len(residue_atoms.intersection(self.phosphate_atoms))
+        scores["phosphate"] = phosphate_match / len(self.phosphate_atoms)
+
+        sugar_match = len(residue_atoms.intersection(self.sugar_atoms))
+        scores["sugar"] = sugar_match / len(self.sugar_atoms)
+
+        nucleobase_atoms = {
+            key: self.nucleobase_heavy_atoms[key] for key in self.nucleobase_heavy_atoms
+        }
+        matches = {
+            key: len(residue_atoms.intersection(nucleobase_atoms[key]))
+            / len(nucleobase_atoms[key])
+            for key in nucleobase_atoms
+        }
+        best_match = max(matches.items(), key=lambda x: x[1])
+        scores["base"] = best_match[1]
+
+        connection_score = 0.0
+        distance_threshold = 2.0
+
+        if "P" in residue_atoms and "O5'" in residue_atoms:
+            p_atom = next(atom for atom in self.atoms if atom.name == "P")
+            o5_atom = next(atom for atom in self.atoms if atom.name == "O5'")
+            if (
+                numpy.linalg.norm(p_atom.coordinates - o5_atom.coordinates)
+                <= distance_threshold
+            ):
+                connection_score += 0.5
+        if "C1'" in residue_atoms:
+            c1_atom = next(atom for atom in self.atoms if atom.name == "C1'")
+            for base_connection in ["N9", "N1"]:
+                if base_connection in residue_atoms:
+                    base_atom = next(
+                        atom for atom in self.atoms if atom.name == base_connection
+                    )
+                    if (
+                        numpy.linalg.norm(c1_atom.coordinates - base_atom.coordinates)
+                        <= distance_threshold
+                    ):
+                        connection_score += 0.5
+                        break
+
+        scores["connections"] = connection_score
+
+        probability = sum(
+            scores[component] * weights[component] for component in scores.keys()
         )
+        return probability > 0.5
 
     @cached_property
     def base_normal_vector(self) -> Optional[numpy.typing.NDArray[numpy.floating]]:
@@ -566,15 +597,14 @@ class Mapping2D3D:
         return self.__generate_bpseq(canonical)
 
     def __generate_bpseq(self, base_pairs):
+        nucleotides = list(filter(lambda r: r.is_nucleotide, self.structure3d.residues))
         result: Dict[int, List] = {}
         residue_map: Dict[Residue3D, int] = {}
         i = 1
 
-        for j, residue in enumerate(
-            filter(lambda r: r.is_nucleotide, self.structure3d.residues)
-        ):
+        for j, residue in enumerate(nucleotides):
             if self.find_gaps and j > 0:
-                previous = self.structure3d.residues[j - 1]
+                previous = nucleotides[j - 1]
 
                 if (
                     not previous.is_connected(residue)