Starting work on Unicycler-scrub

unicycler/unicycler_scrub.py

@@ -0,0 +1,112 @@
+#!/usr/bin/env python3
+"""
+Copyright 2017 Ryan Wick ([email protected])
+https://github.com/rrwick/Unicycler
+
+This module contains the main script for the Unicycler assembly polisher. It is executed when a
+user runs `unicycler_polish` (after installation) or `unicycler_polish-runner.py`.
+
+It uses Pilon, Arrow, Racon and Freebayes (as appropriate for the input reads) to repeatedly polish
+the assembly. The assembly quality is quantified using ALE and polishing will continue until no
+more improvements are possible.
+
+This file is part of Unicycler. Unicycler is free software: you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by the Free Software Foundation,
+either version 3 of the License, or (at your option) any later version. Unicycler is distributed in
+the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
+details. You should have received a copy of the GNU General Public License along with Unicycler. If
+not, see <http://www.gnu.org/licenses/>.
+"""
+
+
+import argparse
+import os
+import sys
+from .misc import MyHelpFormatter, bold, quit_with_error
+from . import log
+
+
+def main():
+    full_command = ' '.join(('"' + x + '"' if ' ' in x else x) for x in sys.argv)
+    args = get_arguments()
+    print_intro_message(args, full_command)
+
+
+
+def get_arguments():
+    parser = argparse.ArgumentParser(description='Unicycler-scrub - read trimming, chimera '
+                                                 'detection and misassembly detection',
+                                     formatter_class=MyHelpFormatter)
+
+    parser.add_argument('-i', '--in', type=str, required=True,
+                        help='These are the reads or assembly to be scrubbed (can be FASTA or '
+                             'FASTQ format')
+    parser.add_argument('-o', '--out', type=str, required=True,
+                        help='The scrubbed reads or assembly will be saved to this file (will '
+                             'have the same format as the --in file format)')
+    parser.add_argument('-r', '--reads', type=str, required=True,
+                        help='These are the reads used to scrub --input (can be FASTA or FASTQ '
+                             'format, can be the same file as --input)')
+    parser.add_argument('-t', '--trim', type=int, default=50,
+                        help='The aggressiveness with which the input will be trimmed (0 to 100, '
+                             'where 0 is no trimming and 100 is very aggressive trimming)')
+    parser.add_argument('-s', '--split', type=int, default=50,
+                        help='The aggressiveness with which the input will be split (0 to 100, '
+                             'where 0 is no splitting and 100 is very aggressive splitting)')
+    args = parser.parse_args()
+
+    if args.trim < 0 or args.trim > 100:
+        quit_with_error('--trim must be between 0 and 100 (inclusive)')
+    if args.split < 0 or args.split > 100:
+        quit_with_error('--split must be between 0 and 100 (inclusive)')
+
+    return args
+
+
+def print_intro_message(args, full_command):
+    log.log_section_header('Starting Unicycler-scrub', single_newline=True)
+
+    log.log_explanation('Unicycler-scrub uses local alignments (from minimap) to trim and/or '
+                        'split an input sequence. Long reads can be used to scrub themselves (by '
+                        'supplying the same reads to --in and --out) to trim the reads and split '
+                        'chimeric reads. Long reads can also be used to check an assembly by '
+                        'searching for misassembly points (by supplying the assembly to --in and '
+                        'the long reads to --out).',
+                        extra_empty_lines_after=0)
+    log.log_explanation('For more information, please see https://github.com/rrwick/Unicycler')
+
+    log.log('Command: ' + bold(full_command))
+    log.log('')
+
+    trim_level_str = '%3d' % args.trim
+    if args.trim == 0.0:
+        trim_level_str += ' (no trimming)'
+    elif args.trim < 20.0:
+        trim_level_str += ' (very conservative trimming)'
+    elif args.trim < 40.0:
+        trim_level_str += ' (conservative trimming)'
+    elif args.trim < 60.0:
+        trim_level_str += ' (moderate trimming)'
+    elif args.trim < 80.0:
+        trim_level_str += ' (aggressive trimming)'
+    else:
+        trim_level_str += ' (very aggressive trimming)'
+
+    split_level_str = '%3d' % args.split
+    if args.split == 0.0:
+        split_level_str += ' (no splitting)'
+    elif args.split < 20.0:
+        split_level_str += ' (very conservative splitting)'
+    elif args.split < 40.0:
+        split_level_str += ' (conservative splitting)'
+    elif args.split < 60.0:
+        split_level_str += ' (moderate splitting)'
+    elif args.split < 80.0:
+        split_level_str += ' (aggressive splitting)'
+    else:
+        split_level_str += ' (very aggressive splitting)'
+
+    log.log('Trim level:  ' + trim_level_str)
+    log.log('Split level: ' +  split_level_str)
+    log.log('')

unicycler_scrub-runner.py

@@ -0,0 +1,21 @@
+#!/usr/bin/env python3
+"""
+Copyright 2017 Ryan Wick ([email protected])
+https://github.com/rrwick/Unicycler
+
+This file is a convenience wrapper for running Unicycler-scrub directly from the source tree. By
+executing `unicycler_scrub-runner.py`, users can run unicycler_scrub without installing Unicycler.
+
+This file is part of Unicycler. Unicycler is free software: you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by the Free Software Foundation,
+either version 3 of the License, or (at your option) any later version. Unicycler is distributed in
+the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
+details. You should have received a copy of the GNU General Public License along with Unicycler. If
+not, see <http://www.gnu.org/licenses/>.
+"""
+
+from unicycler.unicycler_scrub import main
+
+if __name__ == '__main__':
+    main()