opensource/galaxy
3
0
Fork 0
mirror of https://github.com/galaxyproject/galaxy.git synced 2026-09-24 16:30:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

Performance enhancements for Build custom track for UCSC genome browser tool.

Browse Source
This commit is contained in:
Daniel Blankenberg
2012-10-16 15:25:01 -04:00
parent 0168740587
commit 1e55e46201
1 changed files with 31 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/visualization/build_ucsc_custom_track.py
+31 -19
View File
@@ -11,6 +11,22 @@ def stop_err( msg ):
sys.stderr.write( msg )
sys.exit()
FILE_TYPE_TO_TRACK_TYPE = { 'bed': None, 'bedstrict': None, 'bed6': None, 'bed12': None, 'bedgraph':'bedGraph', 'wig':'wiggle_0' }
CHUNK_SIZE = 2**20 #1mb
def get_track_line_is_interval( file_type, name, description, color, visibility ):
if file_type in FILE_TYPE_TO_TRACK_TYPE:
track_type = FILE_TYPE_TO_TRACK_TYPE[ file_type ]
is_interval = False
else:
track_type = None
is_interval = True
track_line = 'track '
if track_type:
track_line += 'type=%s ' % ( track_type )
track_line += 'name="%s" description="%s" color=%s visibility=%s\n' % ( name, description, color, visibility )
return track_line, is_interval
args = sys.argv[1:]
out_fname = args.pop(0)
@@ -22,25 +38,23 @@ first_invalid_line = 0
while args:
# Suck in one dataset worth of arguments
in_fname = args.pop(0)
type = args.pop(0)
file_type = args.pop(0)
colspec = args.pop(0)
name = args.pop(0)
description = args.pop(0)
color = args.pop(0).replace( '-', ',' )
visibility = args.pop(0)
track_line, is_interval = get_track_line_is_interval( file_type, name, description, color, visibility )
# Do the work
if type == "wig":
print >> out, '''track type=wiggle_0 name="%s" description="%s" color=%s visibility=%s''' \
% ( name, description, color, visibility )
for i, line in enumerate( file( in_fname ) ):
print >> out, line,
print >> out
elif type == "bed":
print >> out, '''track name="%s" description="%s" color=%s visibility=%s''' \
% ( name, description, color, visibility )
for i, line in enumerate( file( in_fname ) ):
print >> out, line,
print >> out
in_file = open( in_fname )
out.write( track_line )
if not is_interval:
while True:
chunk = in_file.read( CHUNK_SIZE )
if chunk:
out.write( chunk )
else:
break
else:
# Assume type is interval (don't pass this script anything else!)
try:
@@ -52,29 +66,27 @@ while args:
except:
stop_err( "Columns in interval file invalid for UCSC custom track." )
print >> out, '''track name="%s" description="%s" color=%s visibility=%s''' \
% ( name, description, color, visibility )
i = 0
for i, line in enumerate( file( in_fname ) ):
for i, line in enumerate( in_file ):
line = line.rstrip( '\r\n' )
if line and not line.startswith( '#' ):
fields = line.split( "\t" )
if st > 0:
#strand column is present
try:
print >> out, "%s\t%s\t%s\t%d\t0\t%s" % ( fields[c], fields[s], fields[e], i, fields[st] )
out.write( "%s\t%s\t%s\t%d\t0\t%s\n" % ( fields[c], fields[s], fields[e], i, fields[st] ) )
except:
skipped_lines += 1
if not first_invalid_line:
first_invalid_line = i+1
else:
try:
print >> out, "%s\t%s\t%s" % ( fields[c], fields[s], fields[e] )
out.write( "%s\t%s\t%s\n" % ( fields[c], fields[s], fields[e] ) )
except:
skipped_lines += 1
if not first_invalid_line:
first_invalid_line = i+1
print >> out
out.write( "\n" ) #separating newline
num_tracks += 1
out.close()