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,)
39 cmd_list.append(" ".join(cmd))
42 def pathname_to_run_name(base):
44 Convert a pathname to a base runfolder name
45 handle the case with a trailing /
49 base, name = os.path.split(base)
55 usage = '%prog: [options] runfolder -l 1,2,3 [runfolder -l 5,6 ...]'
57 parser = optparse.OptionParser(usage)
58 parser.add_option('--dry-run', action='store_true',
59 help='print what would be done',
62 parser.add_option('-d', '--dest-dir', dest='dest_dir',
63 help='location to write srf files to',
65 parser.add_option('-s', '--site',
68 parser.add_option('-l', '--lanes', dest='lanes', action="append",
70 help='comma seperated list of lanes to add to srf'
72 parser.add_option('-j', '--jobs', default=1, type='int',
73 help='how many jobs to run simultaneously')
78 def parse_lane_arg(lane_arg):
80 Convert comma sperated list of lane ids to a list of integers
83 for lane in lane_arg.split(','):
86 if lane < 1 or lane > 8:
87 parser.error('Lanes must be in range [1..8]')
90 parser.error('Lane selections must be integers')
93 def main(cmdline=None):
94 parser = make_parser()
95 opts, args = parser.parse_args(cmdline)
98 parser.error('need runfolder arguments')
100 # parse lane arguemnts
102 if len(opts.lanes) == 0:
103 lanes_list = [[1,2,3,4,5,6,7,8]] * len(args)
104 elif len(opts.lanes) == len(args):
105 for lane_arg in opts.lanes:
106 lanes_list.append(parse_lane_arg(lane_arg))
109 "Number of lane arguments must match number of runfolders"
112 # build list of commands
114 for runfolder_path, lanes in zip(args, lanes_list):
115 # normalize paths, either relative to home dirs or current dir
116 runfolder_path = os.path.abspath(runfolder_path)
117 # the last part of the path should be a runfolder name
118 name = pathname_to_run_name(runfolder_path)
119 # so any bustard directories?
120 runs = runfolder.get_runs(runfolder_path)
121 # give up if there are anything other than 1 run
123 print 'ERROR: Too many run directories in %s' %(runfolder_path,)
126 bustard_dir = runs[0].bustard.pathname
127 cmds[bustard_dir] = make_commands(name, lanes, opts.site, opts.dest_dir)
129 print "ERROR: Couldn't find a bustard directory in", runfolder_path
133 for cwd, cmd_list in cmds.items():
136 q = queuecommands.QueueCommands(cmd_list, opts.jobs)
140 for cwd, cmd_list in cmds.items():
143 print 'jobs: ', opts.jobs
147 if __name__ == "__main__":
148 logging.basicConfig(level=logging.DEBUG)
149 sys.exit(main(sys.argv[1:]))