From 5c7a3523a03ffdf304fed1d1100158b249b39021 Mon Sep 17 00:00:00 2001
From: Diane Trout <diane@caltech.edu>
Date: Tue, 22 Apr 2008 00:02:16 +0000
Subject: [PATCH] Extend makebed to lookup metadata out of a copy of the
 fctracker database

---
 gaworkflow/util/makebed.py | 65 ++++++++++++++++++++++++++++++++++++++
 scripts/makebed            | 58 ++++++++++++----------------------
 2 files changed, 86 insertions(+), 37 deletions(-)
 create mode 100755 gaworkflow/util/makebed.py

diff --git a/gaworkflow/util/makebed.py b/gaworkflow/util/makebed.py
new file mode 100755
index 0000000..d2389ba
--- /dev/null
+++ b/gaworkflow/util/makebed.py
@@ -0,0 +1,65 @@
+"""
+Utility functions to make bedfiles.
+"""
+import os
+
+def make_bed_from_eland_stream(instream, outstream, name, description, chromosome_prefix='chr'):
+  """
+  read an eland result file from instream and write a bedfile to outstream
+  """
+  # indexes into fields in eland_result.txt file
+  SEQ = 1
+  CHR = 6
+  START = 7
+  SENSE = 8
+  # map eland_result.txt sense 
+  sense_map = { 'F': '+', 'R': '-'}
+  sense_color = { 'F': '0,0,255', 'R': '255,255,0' }
+  # provide default track names
+  if name is None: name = "track"
+  if description is None: description = "eland result file"
+  bed_header = 'track name="%s" description="%s" visibility=4 itemRgb="ON"'
+  bed_header += os.linesep
+  outstream.write(bed_header % (name, description))
+
+  for line in instream:
+    fields = line.split()
+    # we need more than the CHR field, and it needs to match a chromosome
+    if len(fields) <= CHR or fields[CHR][:3] != chromosome_prefix:
+      continue
+    start = fields[START]
+    stop = int(start) + len(fields[SEQ])
+    chromosome, extension = fields[CHR].split('.')
+    assert extension == "fa"
+    outstream.write('%s %s %d read 0 %s - - %s%s' % (
+      chromosome,
+      start,
+      stop,
+      sense_map[fields[SENSE]], 
+      sense_color[fields[SENSE]],
+      os.linesep  
+    ))
+
+def make_description(database, flowcell_id, lane):
+    """
+    compute a bedfile name and description from the fctracker database
+    """
+    from gaworkflow.util.fctracker import fctracker
+
+    fc = fctracker(database)
+    cells = fc._get_flowcells("where flowcell_id='%s'" % (flowcell_id))
+    if len(cells) != 1:
+      raise RuntimeError("couldn't find flowcell id %s" % (flowcell_id))
+    lane = int(lane)
+    if lane < 1 or lane > 8:
+      raise RuntimeError("flowcells only have lanes 1-8")
+
+    name = "%s-%s" % (flowcell_id, lane)
+
+    cell_id, cell = cells.items()[0]
+    assert cell_id == flowcell_id
+
+    cell_library_id = cell['lane_%d_library_id' %(lane,)]
+    cell_library = cell['lane_%d_library' %(lane,)]
+    description = "%s-%s" % (cell_library['library_name'], cell_library_id)
+    return name, description
diff --git a/scripts/makebed b/scripts/makebed
index 2f1289f..b093835 100755
--- a/scripts/makebed
+++ b/scripts/makebed
@@ -3,42 +3,7 @@ import optparse
 import sys
 import os
 
-def make_bed(instream, outstream, name, description, chromosome_prefix='chr'):
-  """
-  read an eland result file from instream and write a bedfile to outstream
-  """
-  # indexes into fields in eland_result.txt file
-  SEQ = 1
-  CHR = 6
-  START = 7
-  SENSE = 8
-  # map eland_result.txt sense 
-  sense_map = { 'F': '+', 'R': '-'}
-  sense_color = { 'F': '0,0,255', 'R': '255,255,0' }
-  # provide default track names
-  if name is None: name = "track"
-  if description is None: description = "eland result file"
-  bed_header = 'track name="%s" description="%s" visibility=4 itemRgb="ON"'
-  bed_header += os.linesep
-  outstream.write(bed_header % (name, description))
-
-  for line in instream:
-    fields = line.split()
-    # we need more than the CHR field, and it needs to match a chromosome
-    if len(fields) <= CHR or fields[CHR][:3] != chromosome_prefix:
-      continue
-    start = fields[START]
-    stop = int(start) + len(fields[SEQ])
-    chromosome, extension = fields[CHR].split('.')
-    assert extension == "fa"
-    outstream.write('%s %s %d read 0 %s - - %s%s' % (
-      chromosome,
-      start,
-      stop,
-      sense_map[fields[SENSE]], 
-      sense_color[fields[SENSE]],
-      os.linesep  
-    ))
+from gaworkflow.util.makebed import make_bed_from_eland_stream, make_description
 
 def make_parser():
   parser = optparse.OptionParser()
@@ -55,6 +20,15 @@ def make_parser():
   parser.add_option('--chromosome', dest='prefix',
                     help='Set the chromosome prefix name. defaults to "chr"',
                     default='chr')
+  parser.add_option("--database", dest='database',
+                    help="specify location of fctracker database",
+                    default=None)
+  parser.add_option("--flowcell", dest='flowcell',
+                    help="compute name and description from database using flowcell id",
+                    default=None)
+  parser.add_option("--lane", dest='lane',
+                    help='specify which lane to use when retrieving description from database',
+                    default=None)
   return parser
 
 def main(command_line=None):
@@ -84,7 +58,17 @@ def main(command_line=None):
   else:
     outstream = open(options.outname, 'w')
 
-  make_bed(instream, outstream, options.name, options.description, options.prefix)
+  if options.flowcell is not None and options.lane is not None:
+    # get our name/description out of the database
+    name, description = make_description(
+                           options.database, options.flowcell, options.lane
+                        )
+  else:
+    name = options.name
+    description = options.description
+
+  print name, description
+  #make_bed_from_eland_stream(instream, outstream, name, description, options.prefix)
   return 0
 
 if __name__ == "__main__":
-- 
2.30.2