change GCF attribute gcf_id to id

src/nplinker/genomics/gcf.py

@@ -18,29 +18,29 @@ class GCF:
     tools such as BiG-SCAPE and BiG-SLICE.
 
     Attributes:
-        gcf_id: id of the GCF object.
+        id: id of the GCF object.
         bgc_ids: a set of BGC ids that belongs to the GCF.
         bigscape_class: BiG-SCAPE's BGC class.
             BiG-SCAPE's BGC classes are similar to those defined in MiBIG
             but have more categories (7 classes). More details see:
             https://doi.org/10.1038%2Fs41589-019-0400-9.
     """
 
-    def __init__(self, gcf_id: str, /) -> None:
+    def __init__(self, id: str, /) -> None:
         """Initialize the GCF object.
 
         Args:
-            gcf_id: id of the GCF object.
+            id: id of the GCF object.
         """
-        self.gcf_id = gcf_id
+        self.id = id
         self.bgc_ids: set[str] = set()
         self.bigscape_class: str | None = None
         self._bgcs: set[BGC] = set()
         self._strains: StrainCollection = StrainCollection()
 
     def __str__(self) -> str:
         return (
-            f"GCF(id={self.gcf_id}, #BGC_objects={len(self.bgcs)}, #bgc_ids={len(self.bgc_ids)},"
+            f"GCF(id={self.id}, #BGC_objects={len(self.bgcs)}, #bgc_ids={len(self.bgc_ids)},"
             f"#strains={len(self._strains)})."
         )
 
@@ -49,7 +49,7 @@ def __repr__(self) -> str:
 
     def __eq__(self, other) -> bool:
         if isinstance(other, GCF):
-            return self.gcf_id == other.gcf_id and self.bgcs == other.bgcs
+            return self.id == other.id and self.bgcs == other.bgcs
         return NotImplemented
 
     def __hash__(self) -> int:
@@ -58,7 +58,7 @@ def __hash__(self) -> int:
         Note that GCF class is a mutable container. We only hash the GCF id to
         avoid the hash value changes when `self._bgcs` is updated.
         """
-        return hash(self.gcf_id)
+        return hash(self.id)
 
     @property
     def bgcs(self) -> set[BGC]:

src/nplinker/nplinker.py

@@ -270,9 +270,9 @@ def lookup_bgc(self, id):
         """If BGC ``id`` exists, return it. Otherwise return None."""
         return self._bgc_lookup.get(id, None)
 
-    def lookup_gcf(self, gcf_id):
-        """If GCF ``gcf_id`` exists, return it. Otherwise return None."""
-        return self._gcf_lookup.get(gcf_id, None)
+    def lookup_gcf(self, id):
+        """If GCF ``id`` exists, return it. Otherwise return None."""
+        return self._gcf_lookup.get(id, None)
 
     def lookup_spectrum(self, id):
         """If Spectrum ``name`` exists, return it. Otherwise return None."""

src/nplinker/pickler.py

@@ -41,7 +41,7 @@ def persistent_id(self, obj):
         if isinstance(obj, BGC):
             return ("BGC", obj.id)
         elif isinstance(obj, GCF):
-            return ("GCF", obj.gcf_id)
+            return ("GCF", obj.id)
         elif isinstance(obj, Spectrum):
             return ("Spectrum", obj.id)
         elif isinstance(obj, MolecularFamily):

src/nplinker/scoring/metcalf_scoring.py

@@ -387,9 +387,7 @@ def _get_links(
 
         links = []
         if obj_type == "gcf":
-            # TODO CG: the hint and mypy warnings will be gone after renaming all
-            # string ids to `.id`
-            obj_ids = [gcf.gcf_id for gcf in objects]
+            obj_ids = [gcf.id for gcf in objects]
             # spec-gcf
             scores = self.raw_score_spec_gcf.loc[:, obj_ids]
             df = self._get_scores_source_gcf(scores, score_cutoff)

src/nplinker/scoring/utils.py

@@ -21,19 +21,19 @@ def get_presence_gcf_strain(gcfs: Sequence[GCF], strains: StrainCollection) -> p
     """Get the occurence of strains in gcfs.
 
     The occurence is a DataFrame with gcfs as rows and strains as columns,
-    where index is `gcf.gcf_id` and column name is `strain.id`. The values
+    where index is `gcf.id` and column name is `strain.id`. The values
     are 1 if the gcf contains the strain and 0 otherwise.
     """
     df_gcf_strain = pd.DataFrame(
         np.zeros((len(gcfs), len(strains))),
-        index=[gcf.gcf_id for gcf in gcfs],
+        index=[gcf.id for gcf in gcfs],
         columns=[strain.id for strain in strains],
         dtype=int,
     )
     for gcf in gcfs:
         for strain in strains:
             if gcf.has_strain(strain):
-                df_gcf_strain.loc[gcf.gcf_id, strain.id] = 1
+                df_gcf_strain.loc[gcf.id, strain.id] = 1
     return df_gcf_strain
 
 

tests/unit/genomics/test_gcf.py

@@ -23,7 +23,7 @@ def bgc_without_strain():
 def test_init():
     """Test the initialization of GCF."""
     gcf = GCF("1")
-    assert gcf.gcf_id == "1"
+    assert gcf.id == "1"
     assert gcf.bgcs == set()
     assert isinstance(gcf.strains, StrainCollection)
     assert len(gcf.strains) == 0
@@ -77,7 +77,7 @@ def test_add_bgc_wo_strain(bgc_without_strain, caplog):
     """Test add_bgc method with a BGC that does have strain."""
     gcf = GCF("1")
     gcf.add_bgc(bgc_without_strain)
-    assert gcf.gcf_id == "1"
+    assert gcf.id == "1"
     assert gcf.bgcs == {bgc_without_strain}
     assert len(gcf.strains) == 0
     assert "No strain specified for the BGC" in caplog.text

tests/unit/scoring/test_nplinker_scoring.py

@@ -16,7 +16,7 @@ def test_get_links_gcf_standardised_false(npl, mc, gcfs, spectra, mfs, strains_l
     assert isinstance(links, LinkCollection)
     links = links.links  # dict of link values
     assert len(links) == 3
-    assert {i.gcf_id for i in links.keys()} == {"gcf1", "gcf2", "gcf3"}
+    assert {i.id for i in links.keys()} == {"gcf1", "gcf2", "gcf3"}
     assert isinstance(links[gcfs[0]][spectra[0]], ObjectLink)
     assert links[gcfs[0]][spectra[0]].data(mc) == 12
     assert links[gcfs[1]][spectra[0]].data(mc) == -9
@@ -30,7 +30,7 @@ def test_get_links_gcf_standardised_false(npl, mc, gcfs, spectra, mfs, strains_l
     links = npl.get_links(list(gcfs), mc, and_mode=True)
     assert isinstance(links, LinkCollection)
     links = links.links
-    assert {i.gcf_id for i in links.keys()} == {"gcf1", "gcf2", "gcf3"}
+    assert {i.id for i in links.keys()} == {"gcf1", "gcf2", "gcf3"}
     assert isinstance(links[gcfs[0]][spectra[0]], ObjectLink)
     # test scores
     assert links[gcfs[0]][spectra[0]].data(mc) == 12