def main( argv = None ):
"""script main.
parses command line options in sys.argv, unless *argv* is given.
"""
if not argv: argv = sys.argv
# setup command line parser
parser = E.OptionParser( version = "%prog version: $Id: cgat_script_template.py 2871 2010-03-03 10:20:44Z andreas $",
usage = globals()["__doc__"] )
parser.add_option("-a", "--gtf-a", dest="gtf_a", type="string",
help="supply a gtf file - will compress uncompressed files" )
parser.add_option("-b", "--gtf-b", dest = "gtf_b", type = "string",
help="supply a second gtf file - will compress uncompressed files")
parser.add_option("-s", "--scripts-dir", dest = "scripts_dir", type = "string",
help="supply a location for accessory scripts")
parser.add_option( "--no-venn", dest = "no_venn", action="store_true",
help="set if no venn is to be drawn")
## add common options (-h/--help, ...) and parse command line
(options, args) = E.Start( parser, argv = argv )
gtf_files = [options.gtf_a, options.gtf_b]
merged_files = []
prefices = []
E.info("merging gtf files")
for gtf in gtf_files:
if gtf.endswith(".gtf.gz"):
outfile = P.snip(gtf, ".gtf.gz") + ".merged.gtf.gz"
prefices.append(P.snip(gtf, ".gtf.gz"))
merged_files.append(outfile)
statement = '''zcat %s | python %s/gtf2gtf.py --merge-transcripts --log=%s.log | gzip > %s''' % (gtf, options.scripts_dir, outfile, outfile)
P.run()
elif gtf.endswith(".gtf"):
outfile = P.snip(gtf, ".gtf") + ".merged.gtf.gz"
prefices.append(P.snip(gtf,".gtf"))
merged_files.append(outfile)
statement = '''cat %s | python %s/gtf2gtf.py --merge-transcripts --log=%s.log | gzip > %s''' % (gtf, options.scripts_dir, outfile, outfile)
P.run()
else:
raise ValueError("cannot perform merge on %s: is not a gtf file" % gtf)
for prefix in prefices:
if options.gtf_a.find(prefix) != -1:
gtf_a = prefix + ".merged.gtf.gz"
prefix_a = prefix
elif options.gtf_b.find(prefix) != -1:
gtf_b = prefix + ".merged.gtf.gz"
prefix_b = prefix
E.info("intersecting gtf files")
# intersect the resulting merged files
scriptsdir = options.scripts_dir
intersection_out = "_vs_".join([prefix_a, prefix_b]) + ".intersection.gtf.gz"
statement = '''intersectBed -a %(gtf_a)s -b %(gtf_b)s -s -wa
| python %(scriptsdir)s/gtf2gtf.py --merge-transcripts --log=log | gzip > %(intersection_out)s'''
P.run()
if not options.no_venn:
E.info("producing venn diagram for %s vs %s..." % (options.gtf_a, options.gtf_b))
# produce the venn diagram
intersection_file = intersection_out
gtf_a_merged = gtf_a
gtf_b_merged = gtf_b
# create dictionary key
gtf_pair = (gtf_a_merged, gtf_b_merged)
# containers for counts
count_gtf_merged_a = 0
count_gtf_merged_b = 0
count_intersection = 0
# create GTF iterator objects
gtf_iterator_a = GTF.iterator(IOTools.openFile(gtf_pair[0]))
gtf_iterator_b = GTF.iterator(IOTools.openFile(gtf_pair[1]))
gtf_iterator_intersection = GTF.iterator(IOTools.openFile(intersection_file))
# do the counts for each file
E.info("counting entries in %s" % gtf_a)
for entry in gtf_iterator_a:
count_gtf_merged_a += 1
print "counts for gtf-a: ",count_gtf_merged_a
E.info("counting entries in %s" % gtf_b)
for entry in gtf_iterator_b:
count_gtf_merged_b += 1
print "counts for gtf-b: ",count_gtf_merged_b
E.info("counting entries in %s" % intersection_file)
for entry in gtf_iterator_intersection:
count_intersection += 1
print "counts for intersection: ", count_intersection
# this is the important bit - basically take an arbitrary list of numbers to represent the list of lincrna in the refnoncoding set
# then use the intersection count to represent the overlapping section in the lincrna set and add a set of random numbers to this
# set to make up the remaining - non-overlapping set
result = {}
E.info("assembling count lists")
result[gtf_pair] = {"gtf-b" : map(str,xrange(count_gtf_merged_b)) , "gtf-a" : map(str,xrange(count_intersection)) + map(str, [random.random() for i in range(count_intersection,count_gtf_merged_a)] )}
R_source = os.path.join(os.path.abspath(options.scripts_dir), "venn_diagram.R")
R.source(R_source)
prefix_a = prefix_a.replace(".", "_").replace("-", "_")
prefix_b = prefix_b.replace(".", "_").replace("-", "_")
R('''prefix.a <- "%s"''' % prefix_a)
R('''prefix.b <- "%s"''' % prefix_b)
E.info("drawing venn diagram to %s" % (prefix_a + "_vs_" + prefix_b + ".overlap.png"))
R["venn.diagram2"](R.list( A = result[gtf_pair]["gtf-a"], B = result[gtf_pair]["gtf-b"])
, prefix_a + "_vs_" + prefix_b + ".overlap.png"
, **{'cat.cex': 1.5
, 'main.fontfamily': "Arial"
, 'cat.pos':FloatVector((0,0))
, 'cat.fontfamily':"Arial"
, 'main.cex':1.8
, 'height':1000
, 'width':1000
, 'cex':2
, 'fontfamily':"Arial"
, 'lwd':R.c(1,1)
, 'fill':R.c(R.rgb(0,0,0.5,0.5), R.rgb(0.5,0,0,0.5))
, 'category.names':R.c(prefix_a, prefix_b)
, 'margin' : R.c(0.1,0.1,0.1,0.1)
})
## write footer and output benchmark information.
E.Stop()
def draw_r(regions, points, **kwargs):
# initialize the environment
from rpy2.interactive import process_revents
from rpy2.robjects import r
from rpy2.robjects.packages import importr
NA = r("NA")[0]
RGB = lambda rgb: r.rgb(*rgb, maxColorValue=256)
C = lambda seq: r.c(*seq)
OOB = 40
graphics = importr("graphics")
grDevices = importr("grDevices")
process_revents.start()
graphics.par(bg="white")
graphics.split_screen(r.c(2, 1))
graphics.split_screen(r.c(1, 2), screen=2)
graphics.screen(1)
# prepare the regions for plotting
ul, lr = regions.box()
xlim = r.c(ul[0], lr[0])
ylim = r.c(lr[1], ul[1])
# create the main plot window
graphics.plot(r.c(), r.c(), main=regions.name(), type="p", pch="+",
xlim=xlim, ylim=ylim, xlab="", ylab="",
xaxp=r.c(0, lr[0], lr[0]/200), yaxp=r.c(0, lr[1], lr[1]/200),
bg="white")
# plot the polygons in the order given
order = sorted(regions.polys(), key=lambda p: p.area, reverse=True)
for poly in order:
xs, ys = zip(*poly.boundary[0].coords)
color = regions.color(poly.name(), default=NA)
cr, cg, cb = r.col2rgb(color)
rgb = r.rgb(cr, cg, cb, alpha=128, maxColorValue=255)
graphics.polygon(C(xs), C(ys), col=rgb)
# plot the grid
graphics.abline(v=r.c(OOB, lr[0]-OOB), lty=2)
graphics.abline(h=r.seq(0, lr[1], 200), col="lightgray", lty=2)
graphics.abline(v=r.seq(0, lr[0], 200), col="lightgray", lty=2)
# plot the points
xs, ys, names = zip(*[(pt[0].x, pt[0].y, pt[1]) for pt in points])
colors = [RGB(points.Color(name)) for name in names]
graphics.points(C(xs), C(ys), xlab="", ylab="", pch="+", col=C(colors))
# save as a png
if "png" in kwargs and kwargs['png']:
grDevices.dev_print(grDevices.png, file=kwargs['png'], width=lr[0],
height=lr[1])
# derive legend contents: colors, counts, names
tid_counts = {}
uniq_tids = []
for n in names:
if n not in uniq_tids:
tid_counts[n] = 0
uniq_tids.append(n)
tid_counts[n] += 1
uniq_colors = [RGB(points.Color(tid)) for tid in uniq_tids]
uniq_names = [("%d\t%s" % (i, IDs.TileID[i])) for i in uniq_tids]
name_counts = [("%d\t%s: %d" % (k, IDs.TileID[k], v)) for (k,v) in \
tid_counts.items()]
# display the colors legend
legend_args = dict(y_intersp=0.7, cex=0.7)
graphics.screen(3)
graphics.legend("center", title="Tile Colors", legend=C(uniq_names),
col=C(uniq_colors), pch="+", pt_cex=1, **legend_args)
# display the counts legend
graphics.screen(4)
graphics.legend("center", title="Tile Counts", legend=C(name_counts),
**legend_args)
# sleep until the window is closed
while grDevices.dev_list() != r("NULL"):
time.sleep(0.1)
def main(argv=None):
"""script main.
parses command line options in sys.argv, unless *argv* is given.
"""
if not argv:
argv = sys.argv
# setup command line parser
parser = E.OptionParser(
version=
"%prog version: $Id: cgat_script_template.py 2871 2010-03-03 10:20:44Z andreas $",
usage=globals()["__doc__"])
parser.add_option(
"-a",
"--first-gtf-file",
dest="gtf_a",
type="string",
help="supply a gtf file - will compress uncompressed files")
parser.add_option(
"-b",
"--second-gtf-file",
dest="gtf_b",
type="string",
help="supply a second gtf file - will compress uncompressed files")
parser.add_option("-s",
"--scripts-dir",
dest="scripts_dir",
type="string",
help="supply a location for accessory scripts")
parser.add_option("--no-venn",
dest="no_venn",
action="store_true",
help="set if no venn is to be drawn")
# add common options (-h/--help, ...) and parse command line
(options, args) = E.Start(parser, argv=argv)
gtf_files = [options.gtf_a, options.gtf_b]
merged_files = []
prefices = []
E.info("merging gtf files")
for gtf in gtf_files:
if gtf.endswith(".gtf.gz"):
outfile = IOTools.snip(gtf, ".gtf.gz") + ".merged.gtf.gz"
prefices.append(IOTools.snip(gtf, ".gtf.gz"))
merged_files.append(outfile)
statement = '''zcat %s | python %s/gtf2gtf.py --method=merge-transcripts --log=%s.log | gzip > %s''' % (
gtf, options.scripts_dir, outfile, outfile)
P.execute(statement)
elif gtf.endswith(".gtf"):
outfile = IOTools.snip(gtf, ".gtf") + ".merged.gtf.gz"
prefices.append(IOTools.snip(gtf, ".gtf"))
merged_files.append(outfile)
statement = '''cat %s | python %s/gtf2gtf.py --method=merge-transcripts --log=%s.log | gzip > %s''' % (
gtf, options.scripts_dir, outfile, outfile)
E.execute(statement)
else:
raise ValueError("cannot perform merge on %s: is not a gtf file" %
gtf)
for prefix in prefices:
if options.gtf_a.find(prefix) != -1:
gtf_a = prefix + ".merged.gtf.gz"
prefix_a = prefix
elif options.gtf_b.find(prefix) != -1:
gtf_b = prefix + ".merged.gtf.gz"
prefix_b = prefix
E.info("intersecting gtf files")
# intersect the resulting merged files
scriptsdir = options.scripts_dir
intersection_out = "_vs_".join([prefix_a, prefix_b
]) + ".intersection.gtf.gz"
statement = '''intersectBed -a %(gtf_a)s -b %(gtf_b)s -s -wa
| python %(scriptsdir)s/gtf2gtf.py --method=merge-transcripts --log=log | gzip > %(intersection_out)s'''
P.run()
if not options.no_venn:
E.info("producing venn diagram for %s vs %s..." %
(options.gtf_a, options.gtf_b))
# produce the venn diagram
intersection_file = intersection_out
gtf_a_merged = gtf_a
gtf_b_merged = gtf_b
# create dictionary key
gtf_pair = (gtf_a_merged, gtf_b_merged)
# containers for counts
count_gtf_merged_a = 0
count_gtf_merged_b = 0
count_intersection = 0
# create GTF iterator objects
gtf_iterator_a = GTF.iterator(IOTools.openFile(gtf_pair[0]))
gtf_iterator_b = GTF.iterator(IOTools.openFile(gtf_pair[1]))
gtf_iterator_intersection = GTF.iterator(
IOTools.openFile(intersection_file))
# do the counts for each file
E.info("counting entries in %s" % gtf_a)
for entry in gtf_iterator_a:
count_gtf_merged_a += 1
print("counts for gtf-a: ", count_gtf_merged_a)
E.info("counting entries in %s" % gtf_b)
for entry in gtf_iterator_b:
count_gtf_merged_b += 1
print("counts for gtf-b: ", count_gtf_merged_b)
E.info("counting entries in %s" % intersection_file)
for entry in gtf_iterator_intersection:
count_intersection += 1
print("counts for intersection: ", count_intersection)
# this is the important bit - basically take an arbitrary list of numbers to represent the list of lincrna in the refnoncoding set
# then use the intersection count to represent the overlapping section in the lincrna set and add a set of random numbers to this
# set to make up the remaining - non-overlapping set
result = {}
E.info("assembling count lists")
result[gtf_pair] = {
"gtf-b":
list(map(str, range(count_gtf_merged_b))),
"gtf-a":
list(map(str, range(count_intersection))) + list(
map(str, [
random.random()
for i in range(count_intersection, count_gtf_merged_a)
]))
}
R_source = os.path.join(os.path.abspath(options.scripts_dir),
"venn_diagram.R")
R.source(R_source)
prefix_a = prefix_a.replace(".", "_").replace("-", "_")
prefix_b = prefix_b.replace(".", "_").replace("-", "_")
R('''prefix.a <- "%s"''' % prefix_a)
R('''prefix.b <- "%s"''' % prefix_b)
E.info("drawing venn diagram to %s" %
(prefix_a + "_vs_" + prefix_b + ".overlap.png"))
R["venn.diagram2"](R.list(A=result[gtf_pair]["gtf-a"],
B=result[gtf_pair]["gtf-b"]),
prefix_a + "_vs_" + prefix_b + ".overlap.png", **{
'cat.cex':
1.5,
'main.fontfamily':
"Arial",
'cat.pos':
FloatVector((0, 0)),
'cat.fontfamily':
"Arial",
'main.cex':
1.8,
'height':
1000,
'width':
1000,
'cex':
2,
'fontfamily':
"Arial",
'lwd':
R.c(1, 1),
'fill':
R.c(R.rgb(0, 0, 0.5, 0.5),
R.rgb(0.5, 0, 0, 0.5)),
'category.names':
R.c(prefix_a, prefix_b),
'margin':
R.c(0.1, 0.1, 0.1, 0.1)
})
# write footer and output benchmark information.
E.Stop()
