10 from htsworkflow.util import queuecommands
11 from htsworkflow.pipelines import runfolder
13 def make_commands(run_name, lanes, site_name, destdir):
15 make a subprocess-friendly list of command line arguments to run solexa2srf
17 woldlab:080514_HWI-EAS229_0029_20768AAXX:8.srf
20 run_name - most of the file name (run folder name is a good choice)
21 lanes - list of integers corresponding to which lanes to process
22 site_name - name of your "sequencing site" or "Individual"
23 destdir - where to write all the srf files
27 name_prefix = '%s_%%l_%%t_' % (run_name,)
28 destname = '%s_%s_%d.srf' % (site_name, run_name, lane)
29 destdir = os.path.normpath(destdir)
30 dest_path = os.path.join(destdir, destname)
31 seq_pattern = 's_%d_*_seq.txt' % (lane,)
38 #cmd = ['illumina2srf',
43 cmd_list.append(" ".join(cmd))
46 def pathname_to_run_name(base):
48 Convert a pathname to a base runfolder name
49 handle the case with a trailing /
53 base, name = os.path.split(base)
59 usage = '%prog: [options] runfolder -l 1,2,3 [runfolder -l 5,6 ...]'
61 parser = optparse.OptionParser(usage)
62 parser.add_option('--dry-run', action='store_true',
63 help='print what would be done',
66 parser.add_option('-d', '--dest-dir', dest='dest_dir',
67 help='location to write srf files to',
69 parser.add_option('-s', '--site',
72 parser.add_option('-l', '--lanes', dest='lanes', action="append",
74 help='comma seperated list of lanes to add to srf'
76 parser.add_option('-j', '--jobs', default=1, type='int',
77 help='how many jobs to run simultaneously')
79 parser.add_option('-v', '--verbose', dest='verbose',
80 default=False, action='store_true',
81 help='report more about internals (INFO)')
82 parser.add_option('--debug', dest='debug',
83 default=False, action='store_true',
84 help='report even more about internals (DEBUG)')
88 def parse_lane_arg(lane_arg):
90 Convert comma sperated list of lane ids to a list of integers
93 for lane in lane_arg.split(','):
96 if lane < 1 or lane > 8:
97 parser.error('Lanes must be in range [1..8]')
100 parser.error('Lane selections must be integers')
103 def main(cmdline=None):
104 parser = make_parser()
105 opts, args = parser.parse_args(cmdline)
108 logging.basicConfig(level=logging.DEBUG)
110 logging.basicConfig(level=logging.INFO)
112 logging.basicConfig(level=logging.WARNING)
115 parser.error('need runfolder arguments')
117 # parse lane arguemnts
119 if len(opts.lanes) == 0:
120 lanes_list = [[1,2,3,4,5,6,7,8]] * len(args)
121 elif len(opts.lanes) == len(args):
122 for lane_arg in opts.lanes:
123 lanes_list.append(parse_lane_arg(lane_arg))
126 "Number of lane arguments must match number of runfolders"
129 # build list of commands
131 for runfolder_path, lanes in zip(args, lanes_list):
132 # normalize paths, either relative to home dirs or current dir
133 runfolder_path = os.path.abspath(runfolder_path)
134 # the last part of the path should be a runfolder name
135 name = pathname_to_run_name(runfolder_path)
136 # so any bustard directories?
137 runs = runfolder.get_runs(runfolder_path)
138 # give up if there are anything other than 1 run
140 print 'ERROR: Too many run directories in %s' %(runfolder_path,)
143 bustard_dir = runs[0].bustard.pathname
144 cmds[bustard_dir] = make_commands(name, lanes, opts.site, opts.dest_dir)
146 print "ERROR: Couldn't find a bustard directory in", runfolder_path
150 for cwd, cmd_list in cmds.items():
153 q = queuecommands.QueueCommands(cmd_list, opts.jobs)
157 for cwd, cmd_list in cmds.items():
160 print 'jobs: ', opts.jobs
164 if __name__ == "__main__":
165 sys.exit(main(sys.argv[1:]))