chore: fix covalent linkages description

src/plinder/data/utils/annotations/interface_gap.py

@@ -111,6 +111,28 @@ def annotate_interface_gaps(
     larger_radius: float = 8.0,
     use_author_fields: bool = False,
 ) -> dict[str, Any]:
+    """
+    Annotate interface atoms in close proximity to gaps.
+
+    Parameters
+    ----------
+    cif_file: Path
+        System mmcif file
+    protein_chains: list[str] | None = None
+        Protein chains
+    ligand_chains: list[str] | None = None
+        Ligand chains
+    smaller_radius: float = 4.0
+        Smaller radius used for extracting neighboring atoms
+    larger_radius: float = 8.0
+        Larger radius used for extracting neighboring atoms
+    use_author_fields: bool = False
+        Use author-defined fields
+    Returns
+    -------
+    dict[str, Any]
+        Dictionary of interface atoms near gaps
+    """
     pdb_id = cif_file.stem.split("_")[0].replace("pdb_0000", "")
 
     if ".cif" in cif_file.name:

src/plinder/data/utils/annotations/ligand_utils.py

@@ -580,6 +580,18 @@ def annotate_interface_gaps_per_chain(
     interface_proximal_gaps: dict[str, dict[tuple[str, str], dict[str, int]]],
     asym_id: str,
 ) -> tuple[int | None, ...]:
+    """Parse protein-ligand and protein-protein interface atoms in proximity to gaps
+    to count interface atoms within 4 Å and within 8 Å of gaps.
+    This focuses on a specific ligand chain and the interfaces and gaps around it.
+
+    Args:
+        interface_proximal_gaps (dict[str, dict[tuple[str, str], dict[str, int]]]):
+            Interface atoms close to gaps (within 4 Å and  8 Å )
+        asym_id: str
+            Asymmetric unit chain id
+    Returns:
+        tuple[int, int, int, int, int, int]
+    """
     try:
         ppi_atoms_within_4A_of_gap = sum(
             [
@@ -738,7 +750,7 @@ class Ligand(DocBaseModel):
         default_factory=set[str],
         description="Ligand covalent linkages as described in https://mmcif.wwpdb.org/dictionaries/mmcif_pdbx_v50.dic/Categories/struct_conn.html "
         + "with _struct_conn.conn_type_id == 'covale', reported in format "
-        + "{auth_resid}:{resname}{assym_id}{seq_resid}{atom_name}__{auth_resid}:{resname}{assym_id}{seq_resid}{atom_name}",
+        + "{auth_resid}:{resname}:{assym_id}:{seq_resid}:{atom_name}__{auth_resid}:{resname}:{assym_id}:{seq_resid}:{atom_name}",
     )
     neighboring_residues: dict[str, list[int]] = Field(
         default_factory=dict,
@@ -989,7 +1001,7 @@ def from_pli(
         ligand_like_chains: dict[str, str]
             Chain: chain type for other ligand-like chains in the entry
         interface_proximal_gaps: dict[str, dict[tuple[str, str], dict[str, int]]]
-            TODO: document
+            Dictionary of protein-ligand and protein-protein interface atoms in proximity to residue gaps
         all_covalent_dict : dict[str, list[tuple[str, str]]]
             All "covalent" residue in entry as defined by mmcif annotations.
             They types are separated by dictionary key and they include:

src/plinder/data/utils/annotations/mmpdb_utils.py

@@ -264,10 +264,6 @@ def add_mmp_clusters_to_data(
         pocket_similarity_dict
     )
 
-    # Load mmp index
-    # load_mmp_df = pd.read_csv(mmp_index, compression="gzip", sep="\t", header=None)
-    # load_mmp_df.columns = ["SMILES1", "SMILES2", "id1", "id2", "V1>>V2", "CONSTANT"]
-
     # Pad interger lig ids with zeros
     load_mmp_df["id1"] = load_mmp_df.id1.apply(pad_integer_lig_ids_with_zeros)
     load_mmp_df["id2"] = load_mmp_df.id2.apply(pad_integer_lig_ids_with_zeros)