Allow passing either a scoring matrix name or object to Aligner con…

…structor

pyfamsa/_famsa.pyi

@@ -61,7 +61,7 @@ class Aligner:
         n_refinements: int = 100,
         keep_duplicates: bool = False,
         refine: typing.Optional[bool] = None,
-        scoring_matrix: ScoringMatrix = MIQS,
+        scoring_matrix: typing.Union[ScoringMatrix, str, None] = None,
     ) -> None: ...
     def align(self, sequences: typing.Iterable[Sequence]) -> Alignment: ...
     def build_tree(self, sequences: typing.Iterable[Sequence]) -> GuideTree: ...

pyfamsa/_famsa.pyx

@@ -322,7 +322,7 @@ cdef class Aligner:
         int n_refinements=100,
         bool keep_duplicates=False,
         object refine=None,
-        ScoringMatrix scoring_matrix not None=MIQS,
+        object scoring_matrix=None,
     ):
         """__init__(self, *, threads=0, guide_tree="sl", tree_heuristic=None, medoid_threshold=0, n_refinements=100, keep_duplicates=False, refine=None)\n--
 
@@ -350,9 +350,10 @@ cdef class Aligner:
             refine (`bool` or `None`): Set to `True` to force refinement,
                 `False` to disable refinement, or leave as `None` to disable
                 refinement automatically for sets of more than 1000 sequences.
-            scoring_matrix (`~scoring_matrices.ScoringMatrix`): The scoring
-                matrix to use for scoring alignments. By default, the *MIQS*
-                matrix by Yamada & Tomii (2014) is used.
+            scoring_matrix (`~scoring_matrices.ScoringMatrix` or `str`): The 
+                scoring matrix to use for scoring alignments. By default, the 
+                *MIQS* matrix by Yamada & Tomii (2014) is used like in the
+                original FAMSA implementation.
 
         """
         self._params.keepDuplicates = keep_duplicates
@@ -403,9 +404,17 @@ cdef class Aligner:
         else:
             raise ValueError("`n_refinements` argument must be positive")
 
-        if scoring_matrix.alphabet != FAMSA_ALPHABET:
-            raise ValueError(f"invalid scoring matrix alphabet: expected {FAMSA_ALPHABET!r}, got {scoring_matrix.alphabet!r}")
-        self.scoring_matrix = scoring_matrix
+        if scoring_matrix is None:
+            self.scoring_matrix = MIQS
+        elif isinstance(scoring_matrix, str):
+            self.scoring_matrix = ScoringMatrix.from_name(scoring_matrix).shuffle(FAMSA_ALPHABET)
+        elif isinstance(scoring_matrix, ScoringMatrix):
+            if scoring_matrix.alphabet != FAMSA_ALPHABET:
+                raise ValueError(f"invalid scoring matrix alphabet: expected {FAMSA_ALPHABET!r}, got {scoring_matrix.alphabet!r}")
+            self.scoring_matrix = scoring_matrix
+        else:
+            ty = type(scoring_matrix).__name__
+            raise TypeError(f"expected str or ScoringMatrix, found {ty}")
 
     # --- Methods ------------------------------------------------------------
 

pyfamsa/tests/test_aligner.py

@@ -5,6 +5,8 @@
 import textwrap
 import tempfile
 
+from scoring_matrices import ScoringMatrix
+
 from .. import Aligner, Sequence
 from . import fasta, data
 
@@ -44,6 +46,22 @@ def test_adeno_fiber_upgma(self):
         self._test_famsa("adeno_fiber", "upgma", None)
 
 
+class TestAligner(unittest.TestCase):
+
+    def test_init_scoring_matrix_str(self):
+        matrix = ScoringMatrix.from_name("BLOSUM62")
+        aligner = Aligner(scoring_matrix="BLOSUM62")
+        self.assertEqual(aligner.scoring_matrix, matrix)
+
+    def test_init_scoring_matrix_object(self):
+        matrix = ScoringMatrix.from_name("BLOSUM62")
+        aligner = Aligner(scoring_matrix=matrix)
+        self.assertEqual(aligner.scoring_matrix, matrix)
+
+    def test_init_scoring_matrix_error(self):
+        self.assertRaises(TypeError, Aligner, scoring_matrix=1)
+
+
 class TestAlign(unittest.TestCase, _Test):
 
     def test_no_sequence(self):