Mercurial > repos > peterjc > venn_list
changeset 29:96164f587b20 draft
planemo upload for repository https://github.com/peterjc/pico_galaxy/tree/master/tools/venn_list commit 9c506f6a728cd08f2622555e5e55a2710d931e62-dirty
| author | peterjc |
|---|---|
| date | Tue, 28 May 2019 12:43:51 -0400 |
| parents | 7e2b53fd03af |
| children | 0ba14c247dfc |
| files | tools/venn_list/README.rst tools/venn_list/tool_dependencies.xml tools/venn_list/venn_list.py tools/venn_list/venn_list.xml |
| diffstat | 4 files changed, 96 insertions(+), 73 deletions(-) [+] |
line wrap: on
line diff
--- a/tools/venn_list/README.rst Thu May 11 07:03:44 2017 -0400 +++ b/tools/venn_list/README.rst Tue May 28 12:43:51 2019 -0400 @@ -33,8 +33,8 @@ The suggested location is in the Galaxy folder ``tools/plotting`` next to other graph drawing tools, or a dedicated ``tools/venn_list`` directory. -You will also need to install Biopython 1.54 or later, and the R/Bioconductor -package limma. You should already have rpy installed for other Galaxy tools. +You will also need to install Biopython 1.54 or later, matplotlib, and +matplotlib_venn - all available via PyPI or Conda, You will also need to modify the ``tools_conf.xml`` file to tell Galaxy to offer the tool. The suggested location is in the "Graph/Display Data" section. Simply add @@ -79,6 +79,10 @@ - Single quote command line arguments (internal change only) - Python 3 compatible print function. - Capture python script version when run from Galaxy. +v0.0.12 - Internal Python style change. + - Python 3 compatible exception handling. +v0.1.0 - Now uses Python library matplotlib_venn to draw the diagram + instead of R's limma via rpy). Contribution from Frederic Sapet. ======= ====================================================================== @@ -107,7 +111,7 @@ $ planemo shed_upload --tar_only tools/venn_list/ ... - $ tar -tzf shed_upload.tar.gz + $ tar -tzf shed_upload.tar.gz tools/venn_list/README.rst tools/venn_list/tool_dependencies.xml tools/venn_list/venn_list.py
--- a/tools/venn_list/tool_dependencies.xml Thu May 11 07:03:44 2017 -0400 +++ b/tools/venn_list/tool_dependencies.xml Tue May 28 12:43:51 2019 -0400 @@ -1,15 +1,15 @@ -<?xml version="1.0"?> +<?xml version="1.0" ?> <tool_dependency> <package name="galaxy_sequence_utils" version="1.0.1"> - <repository changeset_revision="c38bd3fe9da6" name="package_galaxy_sequence_utils_1_0_1" owner="iuc" toolshed="https://testtoolshed.g2.bx.psu.edu" /> + <repository changeset_revision="c38bd3fe9da6" name="package_galaxy_sequence_utils_1_0_1" owner="iuc" toolshed="https://testtoolshed.g2.bx.psu.edu"/> </package> <package name="limma" version="3.25.3"> - <repository changeset_revision="55850b0ff6a3" name="package_r2_limma_3_25_3" owner="iuc" toolshed="https://testtoolshed.g2.bx.psu.edu" /> + <repository changeset_revision="55850b0ff6a3" name="package_r2_limma_3_25_3" owner="iuc" toolshed="https://testtoolshed.g2.bx.psu.edu"/> </package> <package name="rpy" version="1.0.3"> - <repository changeset_revision="3e11ba53f2a6" name="package_rpy_1_0_3" owner="devteam" toolshed="https://testtoolshed.g2.bx.psu.edu" /> + <repository changeset_revision="3e11ba53f2a6" name="package_rpy_1_0_3" owner="devteam" toolshed="https://testtoolshed.g2.bx.psu.edu"/> </package> <package name="biopython" version="1.67"> - <repository changeset_revision="fc45a61abc2f" name="package_biopython_1_67" owner="biopython" toolshed="https://testtoolshed.g2.bx.psu.edu" /> + <repository changeset_revision="fc45a61abc2f" name="package_biopython_1_67" owner="biopython" toolshed="https://testtoolshed.g2.bx.psu.edu"/> </package> -</tool_dependency> +</tool_dependency> \ No newline at end of file
--- a/tools/venn_list/venn_list.py Thu May 11 07:03:44 2017 -0400 +++ b/tools/venn_list/venn_list.py Tue May 28 12:43:51 2019 -0400 @@ -1,5 +1,9 @@ #!/usr/bin/env python -"""Plot up to 3-way Venn Diagram using R limma vennDiagram (via rpy) + +""" +Plot up to 3-way Venn Diagram. + +It uses the Python libraries matplotlib and matplotlib_venn. This script is copyright 2010-2017 by Peter Cock, The James Hutton Institute (formerly SCRI), UK. All rights reserved. @@ -10,26 +14,25 @@ from __future__ import print_function import sys +from shutil import move if "-v" in sys.argv or "--version" in sys.argv: - print("v0.0.11") + print("v0.1.0") sys.exit(0) try: - import rpy + import matplotlib except ImportError: - sys.exit("Requires the Python library rpy (to call R)") -except RuntimeError, e: - sys.exit("The Python library rpy is not availble for the current R version\n\n%s" % e) + sys.exit("Requires the Python library matplotlib") -try: - rpy.r.library("limma") -except Exception: - sys.exit("Requires the R library limma (for vennDiagram function)") - +matplotlib.use("Agg") +from matplotlib import pyplot as plt if len(sys.argv) - 1 not in [7, 10, 13]: - sys.exit("Expected 7, 10 or 13 arguments (for 1, 2 or 3 sets), not %i" % (len(sys.argv) - 1)) + sys.exit( + "Expected 7, 10 or 13 arguments (for 1, 2 or 3 sets), not %i" + % (len(sys.argv) - 1) + ) all_file, all_type, all_label = sys.argv[1:4] set_data = [] @@ -45,6 +48,7 @@ def load_ids(filename, filetype): + """Load ids from files.""" if filetype == "tabular": for line in open(filename): line = line.rstrip("\n") @@ -57,6 +61,7 @@ elif filetype.startswith("fastq"): # Use the Galaxy library not Biopython to cope with CS from galaxy_utils.sequence.fastq import fastqReader + handle = open(filename, "rU") for record in fastqReader(handle): # The [1:] is because the fastaReader leaves the @ on the identifer. @@ -68,18 +73,21 @@ except ImportError: sys.exit("Require Biopython 1.54 or later (to read SFF files)") # This will read the SFF index block if present (very fast) - for name in index(filename, "sff"): - yield name + for this_name in index(filename, "sff"): + yield this_name else: sys.exit("Unexpected file type %s" % filetype) def load_ids_whitelist(filename, filetype, whitelist): - for name in load_ids(filename, filetype): - if name in whitelist: - yield name + """Check if ids are in whitelist.""" + for single_name in load_ids(filename, filetype): + if single_name in whitelist: + yield single_name else: - sys.exit("Unexpected ID %s in %s file %s" % (name, filetype, filename)) + sys.exit( + "Unexpected ID %s in %s file %s" % (single_name, filetype, filename) + ) if all_file in ["", "-", '""', '"-"']: @@ -98,42 +106,54 @@ for s, (f, t, c) in zip(sets, set_data): print("%i in %s" % (len(s), c)) -# Now call R library to draw simple Venn diagram +names = [one_name for one_file, one_type, one_name in set_data] +lengths = [len(one_set) for one_set in sets] + +if len(sets) == 3: + try: + from matplotlib_venn import venn3_unweighted + except ImportError: + sys.exit("Requires the Python library matplotlib_venn") + venn3_unweighted( + sets, + [ + "{} (Total {})".format(name, length) + for (name, length) in zip(names, lengths) + ], + ) + +if len(sets) == 2: + try: + from matplotlib_venn import venn2_unweighted + except ImportError: + sys.exit("Requires the Python library matplotlib_venn") + venn2_unweighted( + sets, + [ + "{} (Total {})".format(name, length) + for (name, length) in zip(names, lengths) + ], + ) + +# not sure what I am doing here, +# matplotlib_venn does not want to create a single Venn circle +# stick to the old behavior (rpy and Limma) as much as possible +if len(sets) == 1: + try: + from matplotlib_venn import venn2 + except ImportError: + sys.exit("Requires the Python library matplotlib_venn") + venn2((sets[0], set()), [set_data[0][2], ""]) + +plt.title(all_label) +plt.savefig(pdf_file + ".pdf") + +# Format "dat" is not supported. try: - # Create dummy Venn diagram counts object for three groups - cols = 'c("%s")' % '","'.join("Set%i" % (i + 1) for i in range(n)) - rpy.r('groups <- cbind(%s)' % ','.join(['1'] * n)) - rpy.r('colnames(groups) <- %s' % cols) - rpy.r('vc <- vennCounts(groups)') - # Populate the 2^n classes with real counts - # Don't make any assumptions about the class order - # print(rpy.r('vc')) - for index, row in enumerate(rpy.r('vc[,%s]' % cols)): - if isinstance(row, int) or isinstance(row, float): - # Hack for rpy being too clever for single element row - row = [row] - names = all_ids - for wanted, s in zip(row, sets): - if wanted: - names = names.intersection(s) - else: - names = names.difference(s) - rpy.r('vc[%i,"Counts"] <- %i' % (index + 1, len(names))) - # print(rpy.r('vc')) - if n == 1: - # Single circle, don't need to add (Total XXX) line - names = [c for (t, f, c) in set_data] - else: - names = ["%s\n(Total %i)" % (c, len(s)) for s, (f, t, c) in zip(sets, set_data)] - rpy.r.assign("names", names) - rpy.r.assign("colors", ["red", "green", "blue"][:n]) - rpy.r.pdf(pdf_file, 8, 8) - rpy.r("""vennDiagram(vc, include="both", names=names, - main="%s", sub="(Total %i)", - circle.col=colors) - """ % (all_label, len(all_ids))) - rpy.r.dev_off() -except Exception, exc: - sys.exit("%s" % str(exc)) -rpy.r.quit(save="no") + move(pdf_file + ".pdf", pdf_file) +except (OSError, IOError) as error: + sys.exit("Fail to rename file {}".format(str(error))) + +plt.close() + print("Done")
--- a/tools/venn_list/venn_list.xml Thu May 11 07:03:44 2017 -0400 +++ b/tools/venn_list/venn_list.xml Tue May 28 12:43:51 2019 -0400 @@ -1,10 +1,10 @@ -<tool id="venn_list" name="Venn Diagram" version="0.0.11"> +<tool id="venn_list" name="Venn Diagram" version="0.1.0"> <description>from lists</description> <requirements> - <requirement type="package" version="1.0.1">galaxy_sequence_utils</requirement> - <requirement type="package" version="1.0.3">rpy</requirement> - <requirement type="package" version="3.25.3">limma</requirement> + <requirement type="package" version="1.1.3">galaxy_sequence_utils</requirement> <requirement type="package" version="1.67">biopython</requirement> + <requirement type="package" version="2.0.0">matplotlib</requirement> + <requirement type="package" version="0.11.5">matplotlib-venn</requirement> </requirements> <version_command> python $__tool_directory__/venn_list.py --version @@ -126,10 +126,9 @@ **Citations** -The Venn Diagrams are drawn using Gordon Smyth's limma package from -R/Bioconductor, http://www.bioconductor.org/ +The Venn Diagrams are drawn using Konstantin Tretyakov's matplotlib-venn +package, https://pypi.org/project/matplotlib-venn/ -The R library is called from Python via rpy, http://rpy.sourceforge.net/ If you use this Galaxy tool in work leading to a scientific publication please cite: @@ -137,14 +136,14 @@ Peter J.A. Cock, Björn A. Grüning, Konrad Paszkiewicz and Leighton Pritchard (2013). Galaxy tools and workflows for sequence analysis with applications in molecular plant pathology. PeerJ 1:e167 -http://dx.doi.org/10.7717/peerj.167 +https://doi.org/10.7717/peerj.167 This tool uses Biopython to read and write SFF files, so you may also wish to cite the Biopython application note (and Galaxy too of course): Cock et al 2009. Biopython: freely available Python tools for computational molecular biology and bioinformatics. Bioinformatics 25(11) 1422-3. -http://dx.doi.org/10.1093/bioinformatics/btp163 pmid:19304878. +https://doi.org/10.1093/bioinformatics/btp163 pmid:19304878. </help> <citations>