bblean 0.7.6__tar.gz → 0.7.8__tar.gz

{bblean-0.7.6 → bblean-0.7.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bblean
-Version: 0.7.6
+Version: 0.7.8
 Summary: BitBirch-Lean Python package
 Author: The Miranda-Quintana Lab and other BitBirch developers
 Author-email: Ramon Alain Miranda Quintana <quintana@chem.ufl.edu>, Krisztina Zsigmond <kzsigmond@ufl.edu>, Ignacio Pickering <ipickering@ufl.edu>, Kenneth Lopez Perez <klopezperez@chem.ufl.edu>, Miroslav Lzicar <miroslav.lzicar@deepmedchem.com>

{bblean-0.7.6 → bblean-0.7.8}/bblean/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.7.6'
-__version_tuple__ = version_tuple = (0, 7, 6)
+__version__ = version = '0.7.8'
+__version_tuple__ = version_tuple = (0, 7, 8)
 
 __commit_id__ = commit_id = None

{bblean-0.7.6 → bblean-0.7.8}/bblean/cli.py

@@ -76,6 +76,85 @@ def _main(
     pass
 
 
+@app.command("compare", rich_help_panel="Analysis", hidden=True)
+def _compare(
+    clusters_a_path: Annotated[Path, Argument()],
+    clusters_b_path: Annotated[Path, Argument()],
+    ari: Annotated[
+        bool,
+        Option("--ari/--no-ari", help="Adjusted Rand index"),
+    ] = True,
+    ami: Annotated[
+        bool,
+        Option("--ami/--no-ami", help="Adjusted mutual information (slow)"),
+    ] = True,
+    top: Annotated[
+        int,
+        Option("-t", "--top"),
+    ] = 30,
+    use_first_clustering_indices: Annotated[
+        bool,
+        Option("--use-first-clustering-indices/--no-use-first-clustering-indices"),
+    ] = False,
+    verbose: Annotated[
+        bool,
+        Option("-v/-V", "--verbose/--no-verbose"),
+    ] = True,
+) -> None:
+    r"""Compare two clusterings of the same data, using different metrics"""
+    import pickle
+    import numpy as np
+
+    from sklearn.metrics import adjusted_mutual_info_score, adjusted_rand_score
+
+    from bblean._console import get_console
+
+    console = get_console(silent=not verbose)
+
+    if clusters_a_path.is_dir():
+        clusters_a_path = clusters_a_path / "clusters.pkl"
+
+    if clusters_b_path.is_dir():
+        clusters_b_path = clusters_b_path / "clusters.pkl"
+
+    with console.status("[italic]Collecting labels...[/italic]", spinner="dots"):
+        with open(clusters_a_path, mode="rb") as f:
+            clusters = pickle.load(f)
+            total = sum(len(c) for c in clusters)
+            true_labels = np.empty(total, dtype=np.uint64)
+            for i, mol_ids in enumerate(clusters):
+                true_labels[mol_ids] = i
+            idxs_a = np.concatenate(clusters[:top])
+
+        with open(clusters_b_path, mode="rb") as f:
+            clusters = pickle.load(f)
+            total = sum(len(c) for c in clusters)
+            pred_labels = np.empty(total, dtype=np.uint64)
+            for i, mol_ids in enumerate(clusters):
+                pred_labels[mol_ids] = i
+            idxs_b = np.concatenate(clusters[:top])
+    if use_first_clustering_indices:
+        idxs = idxs_a
+    else:
+        idxs = np.unique(np.concatenate((idxs_a, idxs_b)))
+
+    true_labels = true_labels[idxs]
+    pred_labels = pred_labels[idxs]
+
+    timer = Timer()
+    timer.init_timing("total")
+    if ami:
+        with console.status("[italic]Calc. AMI score...[/italic]", spinner="dots"):
+            ami_score = adjusted_mutual_info_score(true_labels, pred_labels)
+        console.print(f"Adjusted Mutual Information (AMI): {ami_score:.4f}")
+
+    if ari:
+        with console.status("[italic]Calc. ARI score...[/italic]", spinner="dots"):
+            ari_score = adjusted_rand_score(true_labels, pred_labels)
+        console.print(f"Adjusted Rand Index (ARI): {ari_score:.4f}")
+    timer.end_timing("total", console, indent=False)
+
+
 @app.command("summary", rich_help_panel="Analysis")
 def _table_summary(
     clusters_path: Annotated[
@@ -1550,6 +1629,14 @@ def _fps_from_smiles(
             help="Whether the smiles file has the format <smiles><tab><field><tab>...",
         ),
     ] = False,
+    replace_dummy_atoms: Annotated[
+        bool,
+        Option(
+            "--replace-dummy/--no-replace-dummy",
+            help="Whether to replace dummy atoms such as [U], [Np], etc. used in synthon spaces",  # noqa
+            hidden=True,
+        ),
+    ] = False,
 ) -> None:
     r"""Generate a `*.npy` fingerprints file from one or more `*.smi` smiles files
 
@@ -1656,7 +1743,7 @@ def _fps_from_smiles(
                 pool.map(
                     create_fp_file,
                     _iter_idxs_and_smiles_batches(
-                        smiles_paths, num_per_batch, tab_separated
+                        smiles_paths, num_per_batch, tab_separated, replace_dummy_atoms
                     ),
                 )
         timer.end_timing("total", console, indent=False)
@@ -1698,7 +1785,7 @@ def _fps_from_smiles(
             pool.starmap(
                 fps_array_filler,
                 _iter_ranges_and_smiles_batches(
-                    smiles_paths, num_per_batch, tab_separated
+                    smiles_paths, num_per_batch, tab_separated, replace_dummy_atoms
                 ),
             )
         fps = np.ndarray((smiles_num, out_dim), dtype=dtype, buffer=fps_shmem.buf)

{bblean-0.7.6 → bblean-0.7.8}/bblean/smiles.py

@@ -31,8 +31,12 @@ def calc_num_smiles(smiles_paths: SmilesPaths) -> int:
     return sum(1 for _ in iter_smiles_from_paths(smiles_paths))
 
 
+# NOTE: replace_dummy is the procedure used in RDKit as of Dec 2024 for Synthon spaces
+# synthons marked with [U], [Np], [Pu], [Am]. These need to be converted
 def iter_smiles_from_paths(
-    smiles_paths: SmilesPaths, tab_separated: bool = False
+    smiles_paths: SmilesPaths,
+    tab_separated: bool = False,
+    replace_dummy_atoms: bool = False,
 ) -> tp.Iterator[str]:
     r"""Iterate over smiles in a sequence of smiles paths
 
@@ -44,10 +48,21 @@ def iter_smiles_from_paths(
     for smi_path in smiles_paths:
         with open(smi_path, mode="rt", encoding="utf-8") as f:
             for smi in f:
-                smi = smi if not tab_separated else smi.split("\t")[0]
+                if tab_separated:
+                    smi = smi.split("\t")[0]
+
                 # Skip headers
                 if smi.lower().strip() == "smiles":
                     continue
+
+                # Replace 'dummy' atoms from synthon spaces
+                if replace_dummy_atoms:
+                    smi = (
+                        smi.replace("[U]", "[1*]")
+                        .replace("[Np]", "[2*]")
+                        .replace("[Pu]", "[3*]")
+                        .replace("[Am]", "[4*]")
+                    )
                 yield smi
 
 
@@ -55,10 +70,12 @@ def _iter_ranges_and_smiles_batches(
     smiles_paths: SmilesPaths,
     num_per_batch: int,
     tab_separated: bool = False,
+    replace_dummy_atoms: bool = False,
 ) -> tp.Iterable[tuple[tuple[int, int], tuple[str, ...]]]:
     start_idx = 0
     for batch in batched(
-        iter_smiles_from_paths(smiles_paths, tab_separated), num_per_batch
+        iter_smiles_from_paths(smiles_paths, tab_separated, replace_dummy_atoms),
+        num_per_batch,
     ):
         size = len(batch)
         end_idx = start_idx + size
@@ -70,7 +87,11 @@ def _iter_idxs_and_smiles_batches(
     smiles_paths: SmilesPaths,
     num_per_batch: int,
     tab_separated: bool = False,
+    replace_dummy_atoms: bool = False,
 ) -> tp.Iterable[tuple[int, tuple[str, ...]]]:
     yield from enumerate(
-        batched(iter_smiles_from_paths(smiles_paths, tab_separated), num_per_batch)
+        batched(
+            iter_smiles_from_paths(smiles_paths, tab_separated, replace_dummy_atoms),
+            num_per_batch,
+        )
     )

{bblean-0.7.6 → bblean-0.7.8}/bblean.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bblean
-Version: 0.7.6
+Version: 0.7.8
 Summary: BitBirch-Lean Python package
 Author: The Miranda-Quintana Lab and other BitBirch developers
 Author-email: Ramon Alain Miranda Quintana <quintana@chem.ufl.edu>, Krisztina Zsigmond <kzsigmond@ufl.edu>, Ignacio Pickering <ipickering@ufl.edu>, Kenneth Lopez Perez <klopezperez@chem.ufl.edu>, Miroslav Lzicar <miroslav.lzicar@deepmedchem.com>