Merge pull request #6 from trstickland/master

Dockerfile and addition of optional commands

.travis.yml

@@ -13,4 +13,4 @@ install:
 env:
   - GENOMETOOLS_PATH='/usr/bin/gt'
 script:
-  - "./run_tests.sh" 
+  - ". ./run_tests.sh" 

Dockerfile

@@ -0,0 +1,31 @@
+FROM ubuntu:bionic
+
+MAINTAINER [email protected]
+
+ENV   BASHRC      /etc/bash.bashrc
+ENV   BUILD_DIR   /gffmunger-build
+ENV   CONF_DIR    /etc/gffmunger
+
+RUN   apt-get update -qq
+RUN   apt-get install -y genometools git python3 python-setuptools python3-biopython python3-pip
+
+# RUN   pip3 install dumper gffutils pyyaml
+
+RUN   grep GENOMETOOLS_PATH ${BASHRC} || bash -c "echo; echo 'export GENOMETOOLS_PATH=\"/usr/bin/gt\"'; echo" >> ${BASHRC}
+
+COPY  . ${BUILD_DIR}
+
+COPY  ./*config.yml  ${CONF_DIR}/
+
+RUN   pip3 install ${BUILD_DIR} && \
+      bash -c "cd ${BUILD_DIR} && . ./run_tests.sh --verbose"
+
+RUN   bash -c "echo; echo 'alias gffmunger=\"gffmunger --config ${CONF_DIR}/gffmunger-config.yml\"'; echo" >> ${BASHRC}
+
+VOLUME /var/data
+
+CMD   echo "Usage:  docker run -v \`pwd\`:/var/data -it <IMAGE_NAME> bash" && \
+      echo "" && \
+      echo "This will place you in a shell with your current working directory accessible as /var/data." && \
+      echo "For help, type" && \
+      echo "  gffmunger --help"

README.md

@@ -1,4 +1,5 @@
 # GFF munger
+
 Munges GFF3 files exported from Chado
 
 [![Build Status](https://travis-ci.org/sanger-pathogens/gffmunger.svg?branch=master)](https://travis-ci.org/sanger-pathogens/gffmunger)   
@@ -18,7 +19,9 @@ Munges GFF3 files exported from Chado
   * [Feedback/Issues](#feedbackissues)
 
 ## Introduction
-Munges GFF3 files exported from Chado (http://www.gmod.org/) database to make them suitable for loading into WebApollo. Currently this involved transferring annotations from polypeptide features to the feature (e.g. mRNA) from which the polypeptide derives.
+Munges GFF3 files exported from Chado (http://www.gmod.org/) database to make them suitable for loading into WebApollo.
+
+Currently supports very few functions, but provides a possible framework for additional functionality.
 
 ## Installation
 There are a number of ways to install GFF munger and details are provided below. If you encounter an issue when installing GFF munger please contact your local system administrator. If you encounter a bug please log it [here](https://github.com/sanger-pathogens/gffmunger/issues) or email us at [email protected].
@@ -36,25 +39,25 @@ Install GFF munger:
 
 `conda install -c bioconda gffmunger`
 
-### Debian/Ubuntu (Trusty/Xenial)
-To install Python3 on Ubuntu, as root run:
-```
-apt-get update -qq
-apt-get install -y git python3 python3-setuptools python3-biopython python3-pip
-pip3 install git+git://github.com/sanger-pathogens/gffmunger.git
-```
 ### Running the tests
-The test can be run from the top level directory:  
+The test can be run from the top level directory:
 ```
 ./run_tests.sh
 ```
-## Usage
+
+## Synopsis
 
 ```
-gffmunger [--input chado_export.gff3.gz] [--fasta chado_export.fasta] [--output webapollo_compatible.gff3] [--quiet|--verbose]
+gffmunger [command1 ... commandN] [--input chado_export.gff3.gz] [--fasta chado_export.fasta] [--output webapollo_compatible.gff3] [--quiet|--verbose]
 ```
 
-Without `--input`, will read from standard input; without `--output`, will write new GFF3 to standard output.  If  `--fasta` is not used, then will attempt to read FASTA data from the input GFF3 file.
+### Commands
+
+*move_polypeptide_annot* (default) transfers annotations from polypeptide features to the feature (e.g. mRNA) from which the polypeptide derives.
+
+### Input/output options
+
+Without `--input`, will read from standard input; without `--output`, will write new GFF3 to standard output.  If  `--fasta` is not used, then will read FASTA data (if present) from the input GFF3 file.
 
 ## License
 GFF munger is free software, licensed under [GPLv3](https://github.com/sanger-pathogens/gffmunger/blob/master/LICENSE).

gffmunger/GFFMunger.py

@@ -14,12 +14,16 @@
 from pyfaidx import Fasta
 
 class GFFMunger:
-   
+
    def __init__(self,options):
+
+      self.known_commands = ['move_polypeptide_annot', 'null']
+
       # CLI options
       if None == options:
          # passing None allow class to be constructed without options being defined; intended for example to permit tests
          # *not* recommended for normal usage, for which the 'gffmunger' script is provided
+         self.commands        = self.known_commands.copy()
          self.verbose         = False
          self.quiet           = False
          self.novalidate      = False
@@ -31,6 +35,7 @@ def __init__(self,options):
          self.gt_path_arg     = None
       else:
          # this should be the normal case
+         self.commands        = options.commands
          self.verbose         = options.verbose
          self.quiet           = options.quiet
          self.novalidate      = options.no_validate
@@ -55,11 +60,18 @@ def setLogLevel(level):
       else:
          setLogLevel(logging.WARNING)
 
+      # check command(s)
+      for n,c in enumerate(self.commands):
+         if c in self.known_commands:
+            self.logger.info('Munge command '+str(n+1)+': '+c)
+         else:
+            raise ValueError('Munge command "'+c+'" not recognized')
+
       # options from configuration file
       config_filename = os.path.join(os.path.dirname(os.path.realpath(__file__)), '..', self.config_file)
       try:
          config_fh   = open(config_filename, 'r');
-         self.config = yaml.load(config_fh)
+         self.config = yaml.safe_load(config_fh)
       except Exception:
          self.logger.critical("Can't read configuration file"+config_filename)
          raise
@@ -148,17 +160,22 @@ def run(self):
             self.import_fasta(self.fasta_file_arg)
          # read GFF3 metadta (and poss. other bits) into text buffer(s)
          self.extract_GFF3_components(self.gff3_input_filename)
-         # transfer annotations from polypeptide features to the feature they derived from
-         self.move_annotations()
+
+         if 'move_polypeptide_annot' in self.commands:
+            self.logger.info('transferring polypeptide feature annotations')
+            # transfer annotations from polypeptide features to the feature they derived from
+            self.move_polypeptide_annotations()
+
          # write new GFF3 to file or stdout
          self.export_gff3()
          # if GFF3 file was written, validate it if required
          if self.output_file is not None and not self.novalidate:
             self.validate_GFF3(self.output_file)
-      # whack temporary files
+
       except Exception:
          self.clean_up()
          raise
+
       self.clean_up()
 
 
@@ -377,7 +394,7 @@ def append(new, buf=''):
 
 
 
-   def move_annotations(self):
+   def move_polypeptide_annotations(self):
       """moves annotations from the polypeptide feature to the feature from which it derives (e.g. mRNA)"""
       num_polypeptide=0
       # this list caches all modified Feature objects

gffmunger/tests/io_tests.py

@@ -49,7 +49,7 @@ def test_000_gff3_with_fasta_io(self):
       with warnings.catch_warnings():
          warnings.filterwarnings("ignore", "unclosed file <_io\.TextIOWrapper", ResourceWarning, "gffutils", 668 )
          try:
-            gffmunger.move_annotations()
+            gffmunger.move_polypeptide_annotations()
          except:
             self.fail("Failed to process valid GFF3 file "+test_gff_file)
       warnings.resetwarnings()
@@ -76,7 +76,7 @@ def test_020_gff3_i0(self):
          self.assertIsInstance(newmunger.import_fasta(), pyfaidx.Fasta)
          self.assertIsInstance(newmunger.import_fasta(test_fasta_file), pyfaidx.Fasta)
          try:
-            newmunger.move_annotations()
+            newmunger.move_polypeptide_annotations()
          except:
             self.fail("Failed to process valid GFF3 file "+test_gff_no_fasta)
       warnings.resetwarnings()
@@ -96,9 +96,9 @@ def test_050_gff_error_handling(self):
          try:
             oldloglevel = yet_another_munger.logger.level
             yet_another_munger.logger.setLevel(logging.CRITICAL)
-            yet_another_munger.move_annotations()
+            yet_another_munger.move_polypeptide_annotations()
             yet_another_munger.logger.setLevel(oldloglevel)
          except AssertionError:
-            self.fail("AssertionError should not be raised by GFFMunger.move_annotations() when processing annotations in "+broken_gff_file)
+            self.fail("AssertionError should not be raised by GFFMunger.move_polypeptide_annotations() when processing annotations in "+broken_gff_file)
       warnings.resetwarnings()
       yet_another_munger.clean_up()

scripts/gffmunger

@@ -15,7 +15,7 @@ version = ''
 try:
 	version = pkg_resources.get_distribution("gffmunger").version
 except pkg_resources.DistributionNotFound:
-	version = '0.0.1'
+	version = '0.0.2'
 
 # looking for this file
 config_filename = 'gffmunger-config.yml'
@@ -28,9 +28,17 @@ if os.path.exists(config_sys_path):
 if config_file_path is None or not os.path.exists(config_file_path):
    config_file_path = os.path.normpath(os.path.join(os.path.dirname(os.path.realpath(__file__)),'..',config_filename))
 
-parser = argparse.ArgumentParser(   description = 'A munger of GFF files.  Proper description to follow.',
-                                    usage       = __file__+' [options]', formatter_class=argparse.ArgumentDefaultsHelpFormatter
+parser = argparse.ArgumentParser(   description       = "Munges GFF files. Use one or more of the following commands:\n"# 80 chars --->|
+                                                      + "  move_polypeptide_annot  transfer annotations from polypeptides to the\n"
+                                                      + "                          feature (e.g. mRNA) they derive from\n"
+                                                      + "  null                    do nothing\n",
+                                    #usage             = __file__+' [command1 .. commandN] [options]',
+                                    #formatter_class   = argparse.ArgumentDefaultsHelpFormatter,
+                                    formatter_class   = argparse.RawTextHelpFormatter
                                     )
+
+parser.add_argument('commands',  default = ['move_polypeptide_annot'],  metavar='command',   type=str,   nargs='*',  help  = "Command(s) defining how the GFF should be munged")
+
 parser.add_argument('--verbose',             action='store_true',    default = False,              help = 'Turn on debugging [%(default)s]')
 parser.add_argument('--quiet', '-q',         action='store_true',    default = False,              help = 'Suppress messages & warnings [%(default)s]')
 parser.add_argument('--no-validate', '-n',   action='store_true',    default = False,              help = 'Do not validate the input GFF3 [%(default)s]')

setup.py

@@ -7,7 +7,7 @@
 def read(fname):
     return open(os.path.join(os.path.dirname(__file__), fname)).read()
 
-version = '0.0.1'
+version = '0.0.2'
 if os.path.exists('VERSION'):
   version = open('VERSION').read().strip()
 
@@ -27,7 +27,8 @@ def read(fname):
     install_requires=[
          'biopython >= 1.68',
          #'pyfastaq >= 3.12.0'
-         'gffutils'
+         'gffutils', # no version requirements known; tested with 0.9
+         'pyyaml'    # no version requirements known; tested with 5.1.1
        ],
     license='GPLv3',
     classifiers=[