Refactor test suite

benchmarks/sequence/align/benchmark_kmer.py

@@ -9,9 +9,8 @@
 
 @pytest.fixture(scope="module")
 def sequence():
-    np.random.seed(0)
     sequence = seq.ProteinSequence()
-    sequence.code = np.random.randint(
+    sequence.code = np.random.default_rng(0).integers(
         len(seq.ProteinSequence.alphabet), size=SEQ_LENGTH
     )
     return sequence

benchmarks/sequence/align/benchmark_multiple.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.sequence.align as align
 import biotite.sequence.io.fasta as fasta
@@ -7,7 +6,7 @@
 
 @pytest.fixture(scope="module")
 def sequences():
-    fasta_file = fasta.FastaFile.read(Path(data_dir("sequence")) / "cas9.fasta")
+    fasta_file = fasta.FastaFile.read(data_dir("sequence") / "cas9.fasta")
     return list(fasta.get_sequences(fasta_file).values())
 
 

benchmarks/sequence/align/benchmark_pairwise.py

@@ -1,5 +1,4 @@
 from functools import partial
-from pathlib import Path
 import pytest
 import biotite.sequence as seq
 import biotite.sequence.align as align
@@ -11,7 +10,7 @@
 
 @pytest.fixture(scope="module")
 def sequences():
-    fasta_file = fasta.FastaFile.read(Path(data_dir("sequence")) / "cas9.fasta")
+    fasta_file = fasta.FastaFile.read(data_dir("sequence") / "cas9.fasta")
     return [seq.ProteinSequence(s) for s in fasta_file.values()]
 
 

benchmarks/sequence/align/benchmark_selector.py

@@ -13,9 +13,8 @@
 
 @pytest.fixture(scope="module")
 def sequence():
-    np.random.seed(0)
     s = seq.NucleotideSequence()
-    s.code = np.random.randint(len(ALPHABET), size=SEQ_LENGTH)
+    s.code = np.random.default_rng(0).integers(len(ALPHABET), size=SEQ_LENGTH)
     return s
 
 

benchmarks/sequence/benchmark_fasta.py

@@ -1,5 +1,3 @@
-import os
-from pathlib import Path
 import pytest
 import biotite.sequence as seq
 import biotite.sequence.io.fasta as fasta
@@ -10,10 +8,11 @@
     ["fasta_path", "seq_type"],
     [
         # Single nucleotide sequence entry
-        (Path(data_dir("sequence")) / "ec_bl21.fasta", seq.NucleotideSequence),
+        (data_dir("sequence") / "ec_bl21.fasta", seq.NucleotideSequence),
         # Multiple protein sequence entries
-        (Path(data_dir("sequence")) / "cas9.fasta", seq.ProteinSequence),
+        (data_dir("sequence") / "cas9.fasta", seq.ProteinSequence),
     ],
+    ids=["ec_bl21.fasta", "cas9.fasta"],
 )
 @pytest.mark.benchmark
 def benchmark_get_sequences(fasta_path, seq_type):
@@ -23,7 +22,5 @@ def benchmark_get_sequences(fasta_path, seq_type):
 
 @pytest.mark.benchmark
 def benchmark_get_a3m_alignments():
-    a3m_file = fasta.FastaFile.read(
-        os.path.join(data_dir("sequence"), "1a00_A_uniref90.a3m")
-    )
+    a3m_file = fasta.FastaFile.read(data_dir("sequence") / "1a00_A_uniref90.a3m")
     fasta.get_a3m_alignments(a3m_file, seq_type=seq.ProteinSequence)

benchmarks/sequence/benchmark_phylo.py

@@ -7,8 +7,7 @@
 
 @pytest.fixture(scope="module")
 def distances():
-    np.random.seed(0)
-    rand = np.random.rand(N, N).astype(np.float32)
+    rand = np.random.default_rng(0).random((N, N)).astype(np.float32)
     distances = (rand + rand.T) / 2
     np.fill_diagonal(distances, 0)
     return distances

benchmarks/structure/benchmark_alphabet.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure.alphabet as strucalph
 import biotite.structure.io.pdbx as pdbx
@@ -9,7 +8,9 @@
 
 @pytest.fixture(scope="module")
 def atoms():
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / f"{PDB_ID}.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(
+        data_dir("structure") / "pdb" / f"{PDB_ID}.bcif"
+    )
     return pdbx.get_structure(pdbx_file, model=1, include_bonds=True)
 
 

benchmarks/structure/benchmark_bonds.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure as struc
 import biotite.structure.info as info
@@ -18,7 +17,9 @@ def load_ccd():
 
 @pytest.fixture(scope="module")
 def atoms():
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / f"{PDB_ID}.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(
+        data_dir("structure") / "pdb" / f"{PDB_ID}.bcif"
+    )
     return pdbx.get_structure(pdbx_file, model=1, include_bonds=True)
 
 

benchmarks/structure/benchmark_celllist.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure as struc
 import biotite.structure.io.pdbx as pdbx
@@ -7,7 +6,7 @@
 
 @pytest.fixture(scope="module")
 def atoms():
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / "1gya.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(data_dir("structure") / "pdb" / "1gya.bcif")
     return pdbx.get_structure(pdbx_file, model=1)
 
 

benchmarks/structure/benchmark_compare.py

@@ -1,5 +1,3 @@
-import itertools
-from pathlib import Path
 import pytest
 import biotite.structure as struc
 import biotite.structure.io.pdbx as pdbx
@@ -9,17 +7,15 @@
 
 @pytest.fixture(scope="module")
 def atoms():
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / "1gya.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(data_dir("structure") / "pdb" / "1gya.bcif")
     atoms = pdbx.get_structure(pdbx_file)
     # Reduce the number of atoms to speed up the benchmark
     return atoms[..., filter_heavy(atoms)]
 
 
 @pytest.mark.benchmark
-@pytest.mark.parametrize(
-    "multi_model, aggregation",
-    itertools.product([False, True], ["all", "chain", "residue", "atom"]),
-)
+@pytest.mark.parametrize("multi_model", [False, True])
+@pytest.mark.parametrize("aggregation", ["all", "chain", "residue", "atom"])
 def benchmark_lddt(atoms, multi_model, aggregation):
     """
     Compute lDDT on different aggregation levels.

benchmarks/structure/benchmark_pdb.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure.info as info
 import biotite.structure.io.pdb as pdb
@@ -15,7 +14,7 @@ def load_ccd():
 
 @pytest.fixture(scope="module")
 def pdb_file_path():
-    return Path(data_dir("structure")) / "1aki.pdb"
+    return data_dir("structure") / "pdb" / "1aki.pdb"
 
 
 @pytest.fixture(scope="module")

benchmarks/structure/benchmark_pdbx.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure.io.pdbx as pdbx
 from tests.util import data_dir
@@ -8,7 +7,7 @@
 
 @pytest.fixture
 def pdbx_file():
-    return pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / f"{PDB_ID}.bcif")
+    return pdbx.BinaryCIFFile.read(data_dir("structure") / "pdb" / f"{PDB_ID}.bcif")
 
 
 @pytest.fixture
@@ -24,7 +23,9 @@ def deserialized_data(pdbx_file):
 
 @pytest.fixture
 def atoms():
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / f"{PDB_ID}.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(
+        data_dir("structure") / "pdb" / f"{PDB_ID}.bcif"
+    )
     return pdbx.get_structure(pdbx_file, model=1, include_bonds=True)
 
 
@@ -34,7 +35,7 @@ def benchmark_deserialize_pdbx(format):
     """
     Deserialize all categories of a CIF or BinaryCIF file.
     """
-    path = Path(data_dir("structure")) / f"{PDB_ID}.{format}"
+    path = data_dir("structure") / "pdb" / f"{PDB_ID}.{format}"
     if format == "cif":
         pdbx_file = pdbx.CIFFile.read(path)
     else:
@@ -76,7 +77,7 @@ def benchmark_get_structure(format, include_bonds):
     """
     Parse a structure from a CIF or BinaryCIF file.
     """
-    path = Path(data_dir("structure")) / f"{PDB_ID}.{format}"
+    path = data_dir("structure") / "pdb" / f"{PDB_ID}.{format}"
     if format == "cif":
         pdbx_file = pdbx.CIFFile.read(path)
     else:
@@ -112,7 +113,7 @@ def benchmark_get_assembly():
     Use BinaryCIF to focus on the performance of the assembly operations, rather than
     file parsing.
     """
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / "1f2n.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(data_dir("structure") / "pdb" / "1f2n.bcif")
     pdbx.get_assembly(pdbx_file, model=1)
 
 

benchmarks/structure/benchmark_sasa.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure as struc
 import biotite.structure.io.pdbx as pdbx
@@ -10,7 +9,7 @@ def atoms():
     """
     A structure that includes hydrogen atoms.
     """
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / "1gya.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(data_dir("structure") / "pdb" / "1gya.bcif")
     return pdbx.get_structure(pdbx_file, model=1, include_bonds=True)
 
 

benchmarks/structure/benchmark_superimpose.py

@@ -1,4 +1,3 @@
-from pathlib import Path
 import pytest
 import biotite.structure as struc
 import biotite.structure.io.pdbx as pdbx
@@ -7,7 +6,7 @@
 
 @pytest.fixture(scope="module")
 def atoms():
-    pdbx_file = pdbx.BinaryCIFFile.read(Path(data_dir("structure")) / "1gya.bcif")
+    pdbx_file = pdbx.BinaryCIFFile.read(data_dir("structure") / "pdb" / "1gya.bcif")
     return pdbx.get_structure(pdbx_file)
 
 

doc/tutorial/structure/trajectories.rst

@@ -31,7 +31,7 @@ These can be extracted as :class:`ndarray` with the
     temp_xtc_file = NamedTemporaryFile("wb", suffix=".xtc", delete=False)
     response = requests.get(
         "https://raw.githubusercontent.com/biotite-dev/biotite/master/"
-        "tests/structure/data/1l2y.xtc"
+        "tests/structure/data/pdb/1l2y.xtc"
     )
     temp_xtc_file.write(response.content)
 

src/biotite/structure/basepairs.py

@@ -372,7 +372,7 @@ def base_pairs_edge(
 
     >>> from os.path import join
     >>> dna_helix = load_structure(
-    ...     join(path_to_structures, "base_pairs", "1qxb.cif")
+    ...     join(path_to_structure_data, "base_pairs", "1qxb.cif")
     ... )
     >>> basepairs = base_pairs(dna_helix)
     >>> interacting_edges = base_pairs_edge(dna_helix, basepairs)
@@ -541,7 +541,7 @@ def base_pairs_glycosidic_bond(
 
     >>> from os.path import join
     >>> dna_helix = load_structure(
-    ...     join(path_to_structures, "base_pairs", "1qxb.cif")
+    ...     join(path_to_structure_data, "base_pairs", "1qxb.cif")
     ... )
     >>> basepairs = base_pairs(dna_helix)
     >>> orientations = base_pairs_glycosidic_bond(dna_helix, basepairs)
@@ -684,7 +684,7 @@ def base_stacking(
 
     >>> from os.path import join
     >>> dna_helix = load_structure(
-    ...     join(path_to_structures, "base_pairs", "1bna.pdb")
+    ...     join(path_to_structure_data, "base_pairs", "1bna.pdb")
     ... )
     >>> stacking_interactions = base_stacking(dna_helix)
     >>> print(dna_helix[stacking_interactions].res_id)
@@ -866,7 +866,7 @@ def base_pairs(
 
     >>> from os.path import join
     >>> dna_helix = load_structure(
-    ...     join(path_to_structures, "base_pairs", "1qxb.cif")
+    ...     join(path_to_structure_data, "base_pairs", "1qxb.cif")
     ... )
     >>> basepairs = base_pairs(dna_helix)
     >>> print(dna_helix[basepairs].res_name)

src/biotite/structure/io/mol/mol.py

@@ -51,7 +51,7 @@ class MOLFile(TextFile):
     --------
 
     >>> from os.path import join
-    >>> mol_file = MOLFile.read(join(path_to_structures, "molecules", "TYR.sdf"))
+    >>> mol_file = MOLFile.read(join(path_to_structure_data, "molecules", "TYR.sdf"))
     >>> atom_array = mol_file.get_structure()
     >>> print(atom_array)
                 0             N         1.320    0.952    1.428

src/biotite/structure/io/mol/sdf.py

@@ -591,7 +591,7 @@ class SDFile(File, MutableMapping[str, SDRecord]):
     Read a SD file and parse the molecular structure:
 
     >>> import os.path
-    >>> file = SDFile.read(os.path.join(path_to_structures, "molecules", "TYR.sdf"))
+    >>> file = SDFile.read(os.path.join(path_to_structure_data, "molecules", "TYR.sdf"))
     >>> molecule = file.record.get_structure()
     >>> print(molecule)
             0             N         1.320    0.952    1.428

src/biotite/structure/io/pdbx/convert.py

@@ -1498,7 +1498,7 @@ def get_component(
 
     >>> import os.path
     >>> file = CIFFile.read(
-    ...     os.path.join(path_to_structures, "molecules", "TYR.cif")
+    ...     os.path.join(path_to_structure_data, "molecules", "TYR.cif")
     ... )
     >>> comp = get_component(file)
     >>> print(comp)

src/biotite/structure/rdf.py

@@ -110,7 +110,7 @@ def rdf(
     ignore the counts for the density for each oxygen to itself.
 
     >>> from os.path import join
-    >>> waterbox = load_structure(join(path_to_structures, "waterbox.gro"))
+    >>> waterbox = load_structure(join(path_to_structure_data, "waterbox.gro"))
     >>> oxygens = waterbox[:, waterbox.atom_name == 'OW']
     >>> bins, g_r = rdf(oxygens, oxygens, bins=49, interval=(0.2, 10), periodic=True)
 

src/biotite/structure/rings.py

@@ -181,7 +181,7 @@ def find_stacking_interactions(
 
     >>> from os.path import join
     >>> dna_helix = load_structure(
-    ...     join(path_to_structures, "base_pairs", "1qxb.cif"), include_bonds=True
+    ...     join(path_to_structure_data, "base_pairs", "1qxb.cif"), include_bonds=True
     ... )
     >>> interactions = find_stacking_interactions(dna_helix)
     >>> for ring_atom_indices_1, ring_atom_indices_2, stacking_type in interactions:

tests/application/test_autodock.py

@@ -2,7 +2,6 @@
 # under the 3-Clause BSD License. Please see 'LICENSE.rst' for further
 # information.
 
-from os.path import join
 import numpy as np
 import pytest
 import biotite.structure as struc
@@ -24,7 +23,7 @@ def test_docking(flexible):
     MAX_DEVIATION = 2.0
 
     # A structure of a straptavidin-biotin complex
-    pdbx_file = pdbx.BinaryCIFFile.read(join(data_dir("application"), "2rtg.bcif"))
+    pdbx_file = pdbx.BinaryCIFFile.read(data_dir("application") / "2rtg.bcif")
     structure = pdbx.get_structure(
         pdbx_file, model=1, extra_fields=["charge"], include_bonds=True
     )

tests/application/test_blast.py

@@ -2,7 +2,6 @@
 # under the 3-Clause BSD License. Please see 'LICENSE.rst' for further
 # information.
 
-import os.path
 import pytest
 import biotite.application.blast as blast
 import biotite.sequence as seq
@@ -83,7 +82,7 @@ def test_tblastn():
 
 
 def test_file_input():
-    path = os.path.join(data_dir("sequence"), "prot.fasta")
+    path = data_dir("sequence") / "prot.fasta"
     blast.BlastWebApp("blastp", path, obey_rules=False)
 
 
@@ -119,9 +118,7 @@ def test_invalid_input():
 @pytest.mark.skipif(cannot_connect_to(BLAST_URL), reason="NCBI BLAST is not available")
 def test_hit_with_selenocysteine():
     # Sequence is taken from issue #344
-    query = seqio.load_sequence(
-        os.path.join(data_dir("sequence"), "selenocysteine.fasta")
-    )
+    query = seqio.load_sequence(data_dir("sequence") / "selenocysteine.fasta")
 
     # Expect hit containing selenocysteine when searching Swiss-Prot
     blast_app = blast.BlastWebApp("blastp", query, "swissprot")