Make biopython an optional dependency

.github/actions/setup-deps/action.yaml

@@ -17,8 +17,6 @@ inputs:
     description: 'use micromamba instead of conda'
     default: false
   # conda-installed min dependencies
-  biopython:
-    default: 'biopython>=1.80'
   codecov:
     default: 'codecov'
   cython:
@@ -54,6 +52,8 @@ inputs:
   tqdm:
     default: 'tqdm>=4.43.0'
   # conda-installed optional dependencies
+  biopython:
+    default: 'biopython>=1.80'
   chemfiles-python:
     default: 'chemfiles-python>=0.9'
   clustalw:
@@ -106,7 +106,6 @@ runs:
       shell: bash -l {0}
       env:
         CONDA_MIN_DEPS: |
-          ${{ inputs.biopython }}
           ${{ inputs.codecov }}
           ${{ inputs.cython }}
           ${{ inputs.fasteners }}
@@ -124,6 +123,7 @@ runs:
           ${{ inputs.threadpoolctl }}
           ${{ inputs.tqdm }}
         CONDA_OPT_DEPS: |
+          ${{ inputs.biopython }}
           ${{ inputs.chemfiles-python }}
           ${{ inputs.clustalw }}
           ${{ inputs.distopia }}

package/MDAnalysis/analysis/align.py

@@ -196,10 +196,14 @@
 
 import numpy as np
 
-import Bio.SeqIO
-import Bio.AlignIO
-import Bio.Align
-import Bio.Align.Applications
+try:
+    import Bio.AlignIO
+    import Bio.Align
+    import Bio.Align.Applications
+except ImportError:
+    HAS_BIOPYTHON = False
+else:
+    HAS_BIOPYTHON = True
 
 import MDAnalysis as mda
 import MDAnalysis.lib.qcprot as qcp
@@ -1018,6 +1022,11 @@
         Tuple of top sequence matching output `('Sequence A', 'Sequence B', score,
         begin, end)`
 
+    Raises
+    ------
+    ImportError
+      If optional dependency Biopython is not available.
+
     Notes
     -----
     If you prefer to work directly with :mod:`Bio.Align` objects then you can
@@ -1057,7 +1066,16 @@
        Replace use of deprecated :func:`Bio.pairwise2.align.globalms` with
        :class:`Bio.Align.PairwiseAligner`.
 
+    .. versionchanged:: 2.7.0
+       Biopython is now an optional dependency which this method requires.
+
     """
+    if not HAS_BIOPYTHON:
+        errmsg = ("The `sequence_alignment` method requires an installation "
+                  "of `Biopython`. Please install `Biopython` to use this "
+                  "method: https://biopython.org/wiki/Download")
+        raise ImportError(errmsg)
+
     aligner = Bio.Align.PairwiseAligner(
         mode="global",
         match_score=match_score,
@@ -1159,15 +1177,29 @@
     :func:`sequence_alignment`, which does not require external
     programs.
 
+
+    Raises
+    ------
+    ImportError
+      If optional dependency Biopython is not available.
+
 
     .. _ClustalW: http://www.clustal.org/
     .. _STAMP: http://www.compbio.dundee.ac.uk/manuals/stamp.4.2/
 
     .. versionchanged:: 1.0.0
        Passing `alnfilename` or `treefilename` as `None` will create a file in
        the current working directory.
+    .. versionchanged:: 2.7.0
+       Biopython is now an optional dependency which this method requires.
 
     """
+    if not HAS_BIOPYTHON:
+        errmsg = ("The `fasta2select` method requires an installation "
+                  "of `Biopython`. Please install `Biopython` to use this "
+                  "method: https://biopython.org/wiki/Download")
+        raise ImportError(errmsg)
+
     if is_aligned:
         logger.info("Using provided alignment {}".format(fastafilename))
         with open(fastafilename) as fasta:

package/MDAnalysis/core/topologyattrs.py

@@ -48,8 +48,14 @@
 import textwrap
 from types import MethodType
 
-import Bio.Seq
-import Bio.SeqRecord
+try:
+    import Bio.Seq
+    import Bio.SeqRecord
+except ImportError:
+    HAS_BIOPYTHON = False
+else:
+    HAS_BIOPYTHON = True
+
 import numpy as np
 
 from ..lib.util import (cached, convert_aa_code, iterable, warn_if_not_unique,
@@ -2810,9 +2816,20 @@
 
         :exc:`TypeError` if an unknown *format* is selected.
 
+        :exc:`ImportError` is the biopython package is not available.
+
 
         .. versionadded:: 0.9.0
+        .. versionchanged:: 2.7.0
+          Biopython is now an optional dependency
         """
+        if not HAS_BIOPYTHON:
+            errmsg = ("The `sequence_alignment` method requires an "
+                      "installation of `Biopython`. Please install "
+                      "`Biopython` to use this method: "
+                      "https://biopython.org/wiki/Download")
+            raise ImportError(errmsg)
+
         formats = ('string', 'Seq', 'SeqRecord')
 
         format = kwargs.pop("format", "SeqRecord")

package/pyproject.toml

@@ -36,7 +36,6 @@ maintainers = [
 requires-python = ">=3.9"
 dependencies = [
     'numpy>=1.22.3',
-    'biopython>=1.80',
     'networkx>=2.0',
     'GridDataFormats>=0.4.0',
     'mmtf-python>=1.0.0',
@@ -86,6 +85,7 @@ extra_formats = [
     "rdkit>=2020.03.1",
 ]
 analysis = [
+    "biopython>=1.80",
     "seaborn",
     "scikit-learn",
     "tidynamics>=1.0.0",

package/setup.py

@@ -594,7 +594,6 @@ def long_description(readme):
 
     install_requires = [
           'numpy>=1.22.3',
-          'biopython>=1.80',
           'networkx>=2.0',
           'GridDataFormats>=0.4.0',
           'mmtf-python>=1.0.0',
@@ -660,6 +659,7 @@ def long_description(readme):
                   'parmed', # ParmEd converter
                   ],
               'analysis': [
+                  'biopython>=1.80',  # sequence generation & alignment
                   'seaborn',  # for annotated heat map and nearest neighbor
                               # plotting in PSA
                   'scikit-learn',  # For clustering and dimensionality

testsuite/MDAnalysisTests/analysis/test_align.py

@@ -24,6 +24,7 @@
 
 import MDAnalysis as mda
 import MDAnalysis.analysis.align as align
+from MDAnalysis.analysis.align import HAS_BIOPYTHON
 import MDAnalysis.analysis.rms as rms
 import os
 import numpy as np
@@ -496,19 +497,27 @@ def test_average_structure_in_memory(self, universe):
         assert avg.filename is None
 
 
-class TestAlignmentProcessing(object):
+class TestAlignmentProcessing:
     seq = FASTA
     error_msg = "selection string has unexpected length"
 
+    @pytest.mark.skipif(HAS_BIOPYTHON, reason='biopython is installed')
+    def test_importerror_biopython(self):
+        errmsg = "The `fasta2select` method requires an installation"
+        with pytest.raises(ImportError, match=errmsg):
+            _ = align.fasta2select(self.seq, is_aligned=True)
+
+    @pytest.mark.skipif(not HAS_BIOPYTHON, reason='requires biopython')
     def test_fasta2select_aligned(self):
         """test align.fasta2select() on aligned FASTA (Issue 112)"""
         sel = align.fasta2select(self.seq, is_aligned=True)
         # length of the output strings, not residues or anything real...
         assert len(sel['reference']) == 30623, self.error_msg
         assert len(sel['mobile']) == 30623, self.error_msg
 
-    @pytest.mark.skipif(executable_not_found("clustalw2"),
-                        reason="Test skipped because clustalw2 executable not found")
+    @pytest.mark.skipif(
+        executable_not_found("clustalw2") or not HAS_BIOPYTHON,
+        reason="Test skipped because clustalw2 executable not found")
     def test_fasta2select_file(self, tmpdir):
         """test align.fasta2select() on a non-aligned FASTA with default
         filenames"""
@@ -518,8 +527,9 @@ def test_fasta2select_file(self, tmpdir):
             assert len(sel['reference']) == 23080, self.error_msg
             assert len(sel['mobile']) == 23090, self.error_msg
 
-    @pytest.mark.skipif(executable_not_found("clustalw2"),
-                        reason="Test skipped because clustalw2 executable not found")
+    @pytest.mark.skipif(
+        executable_not_found("clustalw2") or not HAS_BIOPYTHON,
+        reason="Test skipped because clustalw2 executable not found")
     def test_fasta2select_ClustalW(self, tmpdir):
         """MDAnalysis.analysis.align: test fasta2select() with ClustalW
         (Issue 113)"""
@@ -533,6 +543,7 @@ def test_fasta2select_ClustalW(self, tmpdir):
         assert len(sel['reference']) == 23080, self.error_msg
         assert len(sel['mobile']) == 23090, self.error_msg
 
+    @pytest.mark.skipif(not HAS_BIOPYTHON, reason='requires biopython')
     def test_fasta2select_resids(self, tmpdir):
         """test align.fasta2select() when resids provided (Issue #3124)"""
         resids = [x for x in range(705)]
@@ -554,6 +565,14 @@ def atomgroups():
         mobile = universe.select_atoms("resid 122-159")
         return reference, mobile
 
+    @pytest.mark.skipif(HAS_BIOPYTHON, reason='biopython installed')
+    def test_biopython_import_error(self, atomgroups):
+        ref, mob = atomgroups
+        errmsg = "The `sequence_alignment` method requires an installation of"
+        with pytest.raises(ImportError, match=errmsg):
+            align.sequence_alignment(mob, ref)
+
+    @pytest.mark.skipif(not HAS_BIOPYTHON, reason='requires biopython')
     @pytest.mark.filterwarnings("ignore:`sequence_alignment` is deprecated!")
     def test_sequence_alignment(self, atomgroups):
         reference, mobile = atomgroups
@@ -569,6 +588,7 @@ def test_sequence_alignment(self, atomgroups):
         assert score  == pytest.approx(54.6)
         assert_array_equal([begin, end], [0, reference.n_residues])
 
+    @pytest.mark.skipif(not HAS_BIOPYTHON, reason='requires biopython')
     def test_sequence_alignment_deprecation(self, atomgroups):
         reference, mobile = atomgroups
         wmsg = ("`sequence_alignment` is deprecated!\n"

testsuite/MDAnalysisTests/core/test_residuegroup.py

@@ -26,11 +26,20 @@
 import pytest
 
 import MDAnalysis as mda
-
+from MDAnalysis.core.topologyattrs import HAS_BIOPYTHON
 from MDAnalysisTests.datafiles import PSF, DCD
 
 
-class TestSequence(object):
+@pytest.mark.skipif(HAS_BIOPYTHON, reason="biopython is installed")
+def test_sequence_import_error():
+    p = mda.Universe(PSF, DCD).select_atoms('protein')
+    errmsg = "The `sequence_alignment` method requires an installation"
+    with pytest.raises(ImportError, match=errmsg):
+        _ = p.residues.sequence(format="string")
+
+
+@pytest.mark.skipif(not HAS_BIOPYTHON, reason='requires biopython')
+class TestSequence:
     # all tests are done with the AdK system (PSF and DCD) sequence:
     # http://www.uniprot.org/uniprot/P69441.fasta
     # >sp|P69441|KAD_ECOLI Adenylate kinase OS=Escherichia coli (strain K12) GN=adk PE=1 SV=1