helmkit 0.2.0__tar.gz → 0.3.0__tar.gz

{helmkit-0.2.0 → helmkit-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: helmkit
-Version: 0.2.0
+Version: 0.3.0
 Summary: Parse HELM strings into RDKit molecules
 License-File: LICENSE
 Requires-Python: >=3.11
@@ -17,7 +17,7 @@ A Python library for converting HELM (Hierarchical Editing Language for Macromol
 from helmkit import Molecule
 
 # Create a molecule from a HELM string
-helm_string = "PEPTIDE1{A.R.G}$$$"
+helm_string = "PEPTIDE1{A.R.G}$$$$"
 molecule = Molecule(helm_string)
 
 # Access the RDKit molecule object
@@ -31,7 +31,7 @@ from helmkit import Molecule
 from rdkit.Chem import AllChem, Draw
 
 # Create a simple tripeptide (Ala-Arg-Gly)
-molecule = Molecule("PEPTIDE1{A.R.G}$$$")
+molecule = Molecule("PEPTIDE1{A.R.G}$$$$")
 
 # Generate 2D coordinates for visualization
 AllChem.Compute2DCoords(molecule.mol)
@@ -67,7 +67,7 @@ custom_sdf_path = "/path/to/your/custom_monomers.sdf"
 custom_monomers = load_monomer_library(custom_sdf_path)
 
 # Create molecule with custom monomer data
-molecule = Molecule("PEPTIDE1{A.R.G}$$$", monomer_df=custom_monomers)
+molecule = Molecule("PEPTIDE1{A.R.G}$$$$", monomer_df=custom_monomers)
 ```
 
 ## SDF File Structure Requirements
@@ -110,3 +110,22 @@ Ala
 
 $$$$
 ```
+
+## Parallel Processing of Peptides
+
+For workflows involving a large number of peptides, `helmkit` provides a function to process them in parallel, significantly improving performance.
+
+```python
+from helmkit import load_monomer_library
+from helmkit import load_peptides_in_parallel
+
+# Load your custom monomer data (optional)
+custom_sdf_path = "/path/to/your/custom_monomers.sdf"
+monomer_db = load_monomer_library(custom_sdf_path)
+
+# A list of HELM strings
+helm_strings = ["PEPTIDE1{A.R.G}$$$$", "PEPTIDE1{S.G.T}$$$$"]
+
+# Process peptides in parallel
+molecules = load_peptides_in_parallel(helm_strings, monomer_db)
+```

{helmkit-0.2.0 → helmkit-0.3.0}/README.md

@@ -8,7 +8,7 @@ A Python library for converting HELM (Hierarchical Editing Language for Macromol
 from helmkit import Molecule
 
 # Create a molecule from a HELM string
-helm_string = "PEPTIDE1{A.R.G}$$$"
+helm_string = "PEPTIDE1{A.R.G}$$$$"
 molecule = Molecule(helm_string)
 
 # Access the RDKit molecule object
@@ -22,7 +22,7 @@ from helmkit import Molecule
 from rdkit.Chem import AllChem, Draw
 
 # Create a simple tripeptide (Ala-Arg-Gly)
-molecule = Molecule("PEPTIDE1{A.R.G}$$$")
+molecule = Molecule("PEPTIDE1{A.R.G}$$$$")
 
 # Generate 2D coordinates for visualization
 AllChem.Compute2DCoords(molecule.mol)
@@ -58,7 +58,7 @@ custom_sdf_path = "/path/to/your/custom_monomers.sdf"
 custom_monomers = load_monomer_library(custom_sdf_path)
 
 # Create molecule with custom monomer data
-molecule = Molecule("PEPTIDE1{A.R.G}$$$", monomer_df=custom_monomers)
+molecule = Molecule("PEPTIDE1{A.R.G}$$$$", monomer_df=custom_monomers)
 ```
 
 ## SDF File Structure Requirements
@@ -101,3 +101,22 @@ Ala
 
 $$$$
 ```
+
+## Parallel Processing of Peptides
+
+For workflows involving a large number of peptides, `helmkit` provides a function to process them in parallel, significantly improving performance.
+
+```python
+from helmkit import load_monomer_library
+from helmkit import load_peptides_in_parallel
+
+# Load your custom monomer data (optional)
+custom_sdf_path = "/path/to/your/custom_monomers.sdf"
+monomer_db = load_monomer_library(custom_sdf_path)
+
+# A list of HELM strings
+helm_strings = ["PEPTIDE1{A.R.G}$$$$", "PEPTIDE1{S.G.T}$$$$"]
+
+# Process peptides in parallel
+molecules = load_peptides_in_parallel(helm_strings, monomer_db)
+```

{helmkit-0.2.0 → helmkit-0.3.0}/benchmarks/helmkit-benchmark.py

@@ -14,12 +14,12 @@ def main():
     # (as these do not work with pyPept)
     regex = r"\[[^\]]*[\(\s-][^\]]*\]"
     df = df.filter(pl.col("HELM").str.contains(regex).not_())
+    helms = df["HELM"].to_list()
 
     monomer_db = load_monomer_library(data_dir / "monomers.sdf")
 
     start = time.perf_counter()
-    for row in df.iter_rows(named=True):
-        helm = row["HELM"]
+    for helm in helms:
         Molecule(helm, monomer_db)
     end = time.perf_counter()
     print(f"Processed {df.height} peptides in {end - start:.2f} seconds")

helmkit-0.3.0/benchmarks/helmkit-parallel-benchmark.py

@@ -0,0 +1,29 @@
+import time
+from pathlib import Path
+
+import polars as pl
+from helmkit import load_monomer_library
+from helmkit import load_peptides_in_parallel
+
+
+def main():
+    data_dir = Path(__file__).parent / "data"
+    df = pl.read_csv(data_dir / "peptides.csv")
+
+    # Remove peptides with monomers containing parentheses, spaces or hyphens
+    # (as these do not work with pyPept)
+    regex = r"\[[^\]]*[\(\s-][^\]]*\]"
+    df = df.filter(pl.col("HELM").str.contains(regex).not_())
+    helms = df["HELM"].to_list()
+
+    monomer_db = load_monomer_library(data_dir / "monomers.sdf")
+
+    start = time.perf_counter()
+    load_peptides_in_parallel(helms, monomer_db)
+    end = time.perf_counter()
+    print(f"Processed {df.height} peptides in {end - start:.2f} seconds")
+    print(f"Average time per peptide: {(end - start) / df.height:.6f} seconds")
+
+
+if __name__ == "__main__":
+    main()

{helmkit-0.2.0 → helmkit-0.3.0}/benchmarks/helmkit-with-reload-benchmark.py

@@ -14,10 +14,10 @@ def main():
     # (as these do not work with pyPept)
     regex = r"\[[^\]]*[\(\s-][^\]]*\]"
     df = df.filter(pl.col("HELM").str.contains(regex).not_())
+    helms = df["HELM"].to_list()
 
     start = time.perf_counter()
-    for row in df.iter_rows(named=True):
-        helm = row["HELM"]
+    for helm in helms:
         monomer_db = load_monomer_library(data_dir / "monomers.sdf")
         Molecule(helm, monomer_db)
     end = time.perf_counter()

{helmkit-0.2.0 → helmkit-0.3.0}/benchmarks/pypept-benchmark.py

@@ -15,13 +15,13 @@ def main():
     # (as these do not work with pyPept)
     regex = r"\[[^\]]*[\(\s-][^\]]*\]"
     df = df.filter(pl.col("HELM").str.contains(regex).not_())
+    helms = df["HELM"].to_list()
 
     monomer_lib_dir = str(data_dir.relative_to(Path.cwd()))
     monomer_lib = "monomers.sdf"
 
     start = time.perf_counter()
-    for row in df.iter_rows(named=True):
-        helm = row["HELM"]
+    for helm in helms:
         converter = Converter(helm=helm)
         sequence = Sequence(converter.get_biln(), monomer_lib_dir, monomer_lib)
         Molecule(sequence)

{helmkit-0.2.0 → helmkit-0.3.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "helmkit"
-version = "0.2.0"
+version = "0.3.0"
 description = "Parse HELM strings into RDKit molecules"
 readme = "README.md"
 requires-python = ">=3.11"

helmkit-0.3.0/src/helmkit/__init__.py

@@ -0,0 +1,11 @@
+from .molecule import load_monomer_library
+from .molecule import load_peptides_in_parallel
+from .molecule import Molecule
+from .molecule import SequenceConstants
+
+__all__ = [
+    "load_monomer_library",
+    "load_peptides_in_parallel",
+    "Molecule",
+    "SequenceConstants",
+]

{helmkit-0.2.0 → helmkit-0.3.0}/src/helmkit/molecule.py

@@ -1,4 +1,5 @@
 import copy
+import multiprocessing
 import re
 import warnings
 from functools import lru_cache
@@ -456,3 +457,15 @@ class Molecule:
         """Clean up the molecule by removing dummy atoms and sanitizing."""
         self.mol = Chem.DeleteSubstructs(self.mol, Chem.MolFromSmarts("[#0]"))
         Chem.SanitizeMol(self.mol)
+
+
+def _load_peptide(helm: str, monomer_df: Optional[Dict] = None) -> Molecule:
+    return Molecule(helm, monomer_df)
+
+
+def load_peptides_in_parallel(
+    helms: List[str], monomer_df: Optional[Dict] = None
+) -> List[Molecule]:
+    args = [(helm, monomer_df) for helm in helms]
+    with multiprocessing.Pool() as pool:
+        return pool.starmap(_load_peptide, args)

helmkit-0.3.0/tests/test_parallel_load.py

@@ -0,0 +1,42 @@
+from pathlib import Path
+
+import polars as pl
+from helmkit import load_monomer_library
+from helmkit import load_peptides_in_parallel
+from rdkit import Chem
+from rdkit.Chem import AllChem
+from rdkit.Chem import Draw
+from tqdm import tqdm
+
+
+def main():
+    data_dir = Path(__file__).parent / "data"
+    df = pl.read_csv(data_dir / "peptides.csv")
+    monomer_db = load_monomer_library(str(data_dir / "monomers.sdf"))
+    helms = df["HELM"].to_list()
+
+    molecules_parallel = load_peptides_in_parallel(helms, monomer_db)
+    for m, row in tqdm(
+        zip(molecules_parallel, df.iter_rows(named=True)), total=df.height
+    ):
+        smiles = row["SMILES"]
+        inchi1 = Chem.MolToInchi(m.mol)
+        other = Chem.MolFromSmiles(smiles)
+        inchi2 = Chem.MolToInchi(other)
+        if inchi1 != inchi2:
+            mistmatches_dir = Path(__file__).parent / "mismatches"
+            mistmatches_dir.mkdir(exist_ok=True)
+            AllChem.Compute2DCoords(m.mol, clearConfs=True)
+            Chem.AssignAtomChiralTagsFromStructure(m.mol)
+            AllChem.Compute2DCoords(other, clearConfs=True)
+            Chem.AssignAtomChiralTagsFromStructure(other)
+            # Draw the two molecules to an image
+            img = Draw.MolsToGridImage(
+                [m.mol, other], molsPerRow=2, subImgSize=(800, 800)
+            )
+            img.save(mistmatches_dir / f"{row['ID']}.png")
+            raise ValueError(f"{inchi1} != {inchi2} for row {row}")
+
+
+if __name__ == "__main__":
+    main()

{helmkit-0.2.0 → helmkit-0.3.0}/uv.lock

@@ -113,7 +113,7 @@ wheels = [
 
 [[package]]
 name = "helmkit"
-version = "0.1.0"
+version = "0.3.0"
 source = { editable = "." }
 dependencies = [
     { name = "rdkit" },

helmkit-0.2.0/src/helmkit/__init__.py

@@ -1,5 +0,0 @@
-from .molecule import load_monomer_library
-from .molecule import Molecule
-from .molecule import SequenceConstants
-
-__all__ = ["Molecule", "load_monomer_library", "SequenceConstants"]