Merge pull request #375 from widdowquinn/issue_373

Issue 373 - preliminary merge with multiple fixes to annoyances (warnings/small errors) prior to real fix of #373.

Makefile

@@ -47,7 +47,7 @@ walkthrough: clean_walkthrough
 	pyani report --runs -o C_blochmannia_ANIm/ --formats html excel stdout
 	pyani report --run_results 1 --formats html excel stdout -o C_blochmannia_ANIm/
 	pyani report --run_matrices 1 --formats html excel stdout -o C_blochmannia_ANIm/
-	pyani plot --formats png pdf --method seaborn -o C_blochmannia_ANIm --run_ids 1
+	pyani plot --formats png pdf --method seaborn -o C_blochmannia_ANIm --run_id 1
 	# pyani anib C_blochmannia C_blochmannia_ANIb \
     #     --name "C. blochmannia run 2" \
     #     --labels C_blochmannia/labels.txt --classes C_blochmannia/classes.txt

pyani/pyani_graphics/sns/__init__.py

@@ -37,6 +37,7 @@
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 # THE SOFTWARE.
 """Code to implement Seaborn graphics output for ANI analyses."""
+import warnings
 
 import matplotlib  # pylint: disable=C0411
 import pandas as pd
@@ -106,17 +107,37 @@ def get_clustermap(dfr, params, title=None, annot=True):
     :param title:  str, plot title
     :param annot:  Boolean, add text for cell values?
     """
-    fig = sns.clustermap(
-        dfr,
-        cmap=params.cmap,
-        vmin=params.vmin,
-        vmax=params.vmax,
-        col_colors=params.colorbar,
-        row_colors=params.colorbar,
-        figsize=(params.figsize, params.figsize),
-        linewidths=params.linewidths,
-        annot=annot,
-    )
+
+    # If we do not catch warnings here, then we often get the following warning:
+    #   ClusterWarning: scipy.cluster: The symmetric non-negative hollow
+    #   observation matrix looks suspiciously like an uncondensed distance matrix
+    # The usual solution would be to convert the array with
+    # scipy.spatial.distance.squareform(), but this requires that all values in
+    # the main diagonal are zero, which is not the case for ANI.
+    # As we know this is a (1-distance) matrix, we could just set the diagonal
+    # to zero and fudge it, but this is not a good solution. Instead, we suppress
+    # the warning in a context manager for this function call only, because we
+    # know the warning is not relevant.
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            message=(
+                "scipy.cluster: The symmetric non-negative "
+                "hollow observation matrix looks suspiciously like an "
+                "uncondensed distance matrix"
+            ),
+        )
+        fig = sns.clustermap(
+            dfr,
+            cmap=params.cmap,
+            vmin=params.vmin,
+            vmax=params.vmax,
+            col_colors=params.colorbar,
+            row_colors=params.colorbar,
+            figsize=(params.figsize, params.figsize),
+            linewidths=params.linewidths,
+            annot=annot,
+        )
 
     # add labels for each of the input genomes
     add_labels(fig, params)

pyani/scripts/average_nucleotide_identity.py

@@ -812,7 +812,6 @@ def draw(args: Namespace, filestems: List[str], gformat: str) -> None:
         infilename = fullstem.with_suffix(".tab")
         dfm = pd.read_csv(infilename, index_col=0, sep="\t")
         logger.info("Writing heatmap to %s", outfilename)
-        print(args.labels, args.classes)
         params = pyani_graphics.Params(
             params_mpl(dfm)[filestem],
             pyani_tools.get_labels(args.labels, logger=logger),

pyani/scripts/parsers/plot_parser.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 # (c) The James Hutton Institute 2016-2019
-# (c) University of Strathclyde 2019-2020
+# (c) University of Strathclyde 2019-2022
 # Author: Leighton Pritchard
 #
 # Contact:
@@ -17,7 +17,7 @@
 # The MIT License
 #
 # Copyright (c) 2016-2019 The James Hutton Institute
-# Copyright (c) 2019-2020 University of Strathclyde
+# Copyright (c) 2019-2022 University of Strathclyde
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal

pyani/scripts/parsers/report_parser.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 # (c) The James Hutton Institute 2016-2019
-# (c) University of Strathclyde 2019-2020
+# (c) University of Strathclyde 2019-2022
 # Author: Leighton Pritchard
 #
 # Contact:
@@ -17,7 +17,7 @@
 # The MIT License
 #
 # Copyright (c) 2016-2019 The James Hutton Institute
-# Copyright (c) 2019-2020 University of Strathclyde
+# Copyright (c) 2019-2022 University of Strathclyde
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal

pyani/scripts/subcommands/subcmd_plot.py

@@ -194,6 +194,9 @@ def write_distribution(
             title=f"matrix_{matdata.name}_run{run_id}",
         )
 
+    # Be tidy with matplotlib caches
+    plt.close("all")
+
 
 def write_heatmap(
     run_id: int,

pyani/scripts/subcommands/subcmd_report.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 # (c) The James Hutton Institute 2017-2019
-# (c) University of Strathclyde 2019-2020
+# (c) University of Strathclyde 2019-2022
 # Author: Leighton Pritchard
 #
 # Contact:
@@ -18,7 +18,7 @@
 # The MIT License
 #
 # Copyright (c) 2017-2019 The James Hutton Institute
-# Copyright (c) 2019-2020 University of Strathclyde
+# Copyright (c) 2019-2022 University of Strathclyde
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal
@@ -40,6 +40,7 @@
 """Provides the report subcommand for pyani."""
 
 import logging
+import warnings
 
 from argparse import Namespace
 from typing import List, NamedTuple
@@ -154,12 +155,7 @@ def subcmd_report(args: Namespace) -> int:
             "genome label",
             "genome class",
         ]
-        report(
-            args,
-            session,
-            formats,
-            ReportParams("runs_genomes", statement, headers),
-        )
+        report(args, session, formats, ReportParams("runs_genomes", statement, headers))
 
     # Report table of all runs in which a genome is involved
     if args.show_genomes_runs:
@@ -196,12 +192,7 @@ def subcmd_report(args: Namespace) -> int:
             "method",
             "date run",
         ]
-        report(
-            args,
-            session,
-            formats,
-            ReportParams("genomes_runs", statement, headers),
-        )
+        report(args, session, formats, ReportParams("genomes_runs", statement, headers))
 
     # Report table of comparison results for the indicated runs
     if args.run_results:
@@ -234,6 +225,7 @@ def subcmd_report(args: Namespace) -> int:
                 .filter(Run.run_id == run_id)
                 .statement
             )
+            logger.debug("Results query: %s", statement)
             headers = [
                 "Comparison ID",
                 "Query ID",
@@ -300,12 +292,7 @@ def subcmd_report(args: Namespace) -> int:
     return 0
 
 
-def report(
-    args: Namespace,
-    session,
-    formats: List[str],
-    params: ReportParams,
-) -> None:
+def report(args: Namespace, session, formats: List[str], params: ReportParams) -> None:
     """Write tabular report of pyani runs from database.
 
     :param args:  Namespace of command-line arguments
@@ -319,7 +306,47 @@ def report(
     logger.debug(
         "Writing table of pyani %s from the database to %s.*", params.name, outfname
     )
-    data = pd.read_sql(params.statement, session.bind)
+
+    # With newer versions of SQLAlchemy, Pandas may throw a warning due to the composition
+    # of our statement including a Cartesian product, even though this is what we want:
+    #   SAWarning: SELECT statement has a cartesian product between FROM element(s)
+    #   "runs" and FROM element "genome_query".  Apply join condition(s) between each
+    #   element to resolve.
+    # We could use SQLAlchemy's true() function to force the join condition, but this has
+    # to be done from within Pandas, and is an issue for them to fix.
+    # We suppress the warning, instead.
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            message=(
+                "SELECT statement has a cartesian product between FROM "
+                'element\\(s\\) "runs" and FROM element "genome_query"'
+            ),
+        )
+        warnings.filterwarnings(
+            "ignore",
+            message=(
+                "SELECT statement has a cartesian product between FROM "
+                'element\\(s\\) "runs" and FROM element "comparisons"'
+            ),
+        )
+        warnings.filterwarnings(
+            "ignore",
+            message=(
+                "SELECT statement has a cartesian product between FROM "
+                'element\\(s\\) "runs" and FROM element "genome_subject"'
+            ),
+        )
+        warnings.filterwarnings(
+            "ignore",
+            message=(
+                "SELECT statement has a cartesian product between FROM "
+                'element\\(s\\) "genome_query", "genome_subject", "comparisons" '
+                'and FROM element "runs"'
+            ),
+        )
+        data = pd.read_sql(params.statement, session.bind)
+
     data.columns = params.headers
     pyani_report.write_dbtable(data, outfname, formats)
 

requirements.txt

@@ -10,5 +10,5 @@ Pillow
 PyQt5
 scipy
 seaborn
-sqlalchemy==1.3.10
+sqlalchemy
 tqdm

tests/test_anib.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 # (c) The James Hutton Institute 2017-2019
-# (c) University of Strathclyde 2019-2020
+# (c) University of Strathclyde 2019-2022
 # Author: Leighton Pritchard
 #
 # Contact: leighton.pritchard@strath.ac.uk
@@ -17,7 +17,7 @@
 # The MIT License
 #
 # Copyright (c) 2017-2019 The James Hutton Institute
-# Copyright (c) 2019-2020 University of Strathclyde
+# Copyright (c) 2019-2022 University of Strathclyde
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal
@@ -49,7 +49,6 @@
 import pandas as pd
 import pytest  # noqa: F401  # pylint: disable=unused-import
 import unittest
-import unittest.mock as mock
 
 from pandas.testing import assert_frame_equal
 
@@ -367,8 +366,8 @@ def test_parse_legacy_blastdir(anib_output_dir):
         anib_output_dir.legacyblastdir, orglengths, fraglengths, mode="ANIblastall"
     )
     assert_frame_equal(
-        result.percentage_identity.sort_index(1).sort_index(),
-        anib_output_dir.legacyblastresult.sort_index(1).sort_index(),
+        result.percentage_identity.sort_index(axis=1).sort_index(),
+        anib_output_dir.legacyblastresult.sort_index(axis=1).sort_index(),
     )
 
 
@@ -380,8 +379,8 @@ def test_parse_blastdir(anib_output_dir):
         anib_output_dir.blastdir, orglengths, fraglengths, mode="ANIb"
     )
     assert_frame_equal(
-        result.percentage_identity.sort_index(1).sort_index(),
-        anib_output_dir.blastresult.sort_index(1).sort_index(),
+        result.percentage_identity.sort_index(axis=1).sort_index(),
+        anib_output_dir.blastresult.sort_index(axis=1).sort_index(),
     )
 
 

tests/test_anim.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 # (c) The James Hutton Institute 2017-2019
-# (c) University of Strathclyde 2019-2021
+# (c) University of Strathclyde 2019-2022
 # Author: Leighton Pritchard
 #
 # Contact:
@@ -18,7 +18,7 @@
 # The MIT License
 #
 # Copyright (c) 2017-2019 The James Hutton Institute
-# Copyright (c) 2019-2021 University of Strathclyde
+# Copyright (c) 2019-2022 University of Strathclyde
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal
@@ -43,15 +43,15 @@
 
 pytest -v
 """
-import sys
+
 from pathlib import Path
 from typing import List, NamedTuple, Tuple
 
 import pandas as pd
 import pytest
 import unittest
 
-from pandas.util.testing import assert_frame_equal
+from pandas.testing import assert_frame_equal
 
 from pyani import anim, pyani_files
 
@@ -208,8 +208,8 @@ def test_deltadir_parsing(delta_output_dir):
     orglengths = pyani_files.get_sequence_lengths(seqfiles)
     result = anim.process_deltadir(delta_output_dir.deltadir, orglengths)
     assert_frame_equal(
-        result.percentage_identity.sort_index(1).sort_index(),
-        delta_output_dir.deltaresult.sort_index(1).sort_index(),
+        result.percentage_identity.sort_index(axis=1).sort_index(),
+        delta_output_dir.deltaresult.sort_index(axis=1).sort_index(),
     )
 
 
@@ -251,17 +251,14 @@ def test_mummer_single(tmp_path, path_file_two):
         path_file_two[0], path_file_two[1], outdir=tmp_path
     )
     dir_nucmer = tmp_path / "nucmer_output"
+    outprefix = (
+        dir_nucmer
+        / str(path_file_two[0].stem)
+        / str(path_file_two[0].stem + "_vs_" + path_file_two[1].stem)
+    )
     expected = (
-        (
-            "nucmer --mum -p "
-            f"{dir_nucmer / str(path_file_two[0].stem) / str(path_file_two[0].stem + '_vs_' + path_file_two[1].stem)} "
-            f"{path_file_two[0]} {path_file_two[1]}"
-        ),
-        (
-            "delta_filter_wrapper.py delta-filter -1 "
-            f"{dir_nucmer / str(path_file_two[0].stem ) / str(path_file_two[0].stem + '_vs_' + path_file_two[1].stem + '.delta')} "
-            f"{dir_nucmer / str(path_file_two[0].stem ) / str(path_file_two[0].stem + '_vs_' + path_file_two[1].stem + '.filter')}"
-        ),
+        f"nucmer --mum -p {outprefix} {path_file_two[0]} {path_file_two[1]}",
+        f"delta_filter_wrapper.py delta-filter -1 {outprefix}.delta {outprefix}.filter",
     )
     assert cmds == expected
 

tests/test_subcmd_05_report.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 # (c) The James Hutton Institute 2017-2019
-# (c) The University of Strathclude 2019
+# (c) The University of Strathclude 2019-2022
 # Author: Leighton Pritchard
 #
 # Contact:
@@ -19,7 +19,7 @@
 # The MIT License
 #
 # Copyright (c) 2017-2018 The James Hutton Institute
-# (c) The University of Strathclude 2019
+# (c) The University of Strathclude 2019-2022
 #
 # Permission is hereby granted, free of charge, to any person obtaining a copy
 # of this software and associated documentation files (the "Software"), to deal