def plotASProfile(tcc, cName, directory = None, min = 0, extra = "0"):
if not directory:
fN = extra + '.' + tcc + '.png'
else:
fN = directory + '/' + extra + '.' + tcc + '.png'
#Get S Profile
tccStretch = cgPeaks.stretch(tcc, cName)
highest = tccStretch.getHighestLevel()
if highest < min:
return 0
sortedX = tccStretch.profile.keys()
sortedX.sort()
sortedY = []
for X in sortedX:
sortedY.append(tccStretch.profile[X])
#Get AS Profile
chr, strand, start, end = tcc.strip().split(':')
if strand == '1':
strand = '-1'
else:
strand = '1'
tcc = cg.makeTcc(chr, strand, start, end)
tccStretchAS = cgPeaks.stretch(tcc, cName)
highest = tccStretchAS.getHighestLevel()
if highest < min:
return 0 #AS can have minimum I guess...
sortedXAS = tccStretchAS.profile.keys()
sortedXAS.sort()
sortedYAS = []
for X in sortedXAS:
sortedYAS.append(tccStretchAS.profile[X])
#Plot them
gDevice = importr('grDevices')
gDevice.png(file=fN, width=1680, height=1050)
r('split.screen(c(2,1))')
r('screen(1)')
r.plot(sortedX, sortedY, xlab = "Coordinates", ylab = "(Syn) Expression Level" )
r.lines(sortedX, sortedY, type = "b")
r('screen(2)')
r.plot(sortedXAS, sortedYAS, xlab = "Coordinates", ylab = "(Anti) Expression Level")
r.lines(sortedXAS, sortedYAS, type = "b")
gDevice.dev_off()
def val(self):
""" Estimate value functions with b-splines and compare """
new_data = pd.DataFrame({'OverallRank': np.linspace(1, 194, 1000)})
fit_a = self.spline_est(self.policy_a['value'], new_data)
fit_b = self.spline_est(self.policy_b['value'], new_data)
r.pdf(os.path.join(os.path.dirname(self.out_dir), 'value.pdf'))
r.plot(new_data['OverallRank'], fit_a, type='l', xlab='Rank_M',
ylab='V(Rank)')
r.lines(new_data['OverallRank'], fit_b, col='red')
r.points(self.policy_a['value']['OverallRank'],
self.policy_a['value']['val'],
col='black')
r.points(self.policy_b['value']['OverallRank'],
self.policy_b['value']['val'],
col='red')
r.legend('topright', np.array(['No Info', 'Info']),
lty=np.array([1, 1]), col=np.array(['black', 'red']))
r('dev.off()')
diff = np.array(fit_b) - np.array(fit_a)
r.pdf(os.path.join(os.path.dirname(self.out_dir), 'value_diff.pdf'))
r.plot(new_data['OverallRank'], diff, type='l', xlab='Rank',
ylab='V(Rank|info=1) - V(Rank|info=0)')
r.abline(h=0, lty=2)
r('dev.off()')
diff = (np.array(fit_b) - np.array(fit_a)) / np.array(fit_a)
r.pdf(os.path.join(os.path.dirname(self.out_dir),
'value_percent_diff.pdf'))
r.plot(new_data['OverallRank'], diff, type='l', xlab='Rank',
ylab='(V(Rank|info=1) - V(Rank|info=0)) / V(Rank|info=0)')
r.abline(h=0, lty=2)
r('dev.off()')
data_path = dirname(dirname(__file__))
data_path = join(data_path, 'data', 'lawData.csv')
data = pd.read_csv(data_path)
new_data = deepcopy(data.loc[data['year'] == 2013, 'OverallRank'])
#new_data = np.concatenate((
# new_data, np.zeros(lc.N_SCHOOLS - len(new_data))
#))
new_data = pd.DataFrame({'OverallRank': np.array(new_data)})
fit_a = self.spline_est(self.policy_a['value'], new_data)
fit_b = self.spline_est(self.policy_b['value'], new_data)
diff = np.sum(np.array(fit_b) - np.array(fit_a))
pdiff = diff / np.sum(fit_a)
print(" - Change in Producer Surplus: {0}".format(diff))
print(" - Percent change in Producer Surplus: {0}".format(pdiff))
return diff
def plotSmallDeg(tcc, smallCName, degCName, outDir = None, description = "None", nameNum = "0"):
if not outDir:
fN = nameNum + "." + tcc + '.png'
else:
fN = outDir + '/' + nameNum + "." + tcc + '.png'
#Get deg Profile
tccStretch = cgPeaks.stretch(tcc, degCName)
sortedX = tccStretch.profile.keys()
sortedX.sort()
sortedY = []
for X in sortedX:
sortedY.append(tccStretch.profile[X])
#Get small
tccStretchSmall = cgPeaks.stretch(tcc, smallCName)
sortedXAS = tccStretchSmall.profile.keys()
sortedXAS.sort()
sortedYAS = []
for X in sortedXAS:
sortedYAS.append(tccStretchSmall.profile[X])
#Plot them
gDevice = importr('grDevices')
gDevice.png(file=fN, width=1680, height=1050)
r('split.screen(c(2,1))')
r('screen(1)')
r.plot(sortedX, sortedY, xlab = "Coordinates", ylab = "Degradome Expression" )
r.lines(sortedX, sortedY, type = "b")
r('screen(2)')
r.plot(sortedXAS, sortedYAS, xlab = description, ylab = "Small Expression")
r.lines(sortedXAS, sortedYAS, type = "b")
gDevice.dev_off()
def plotProfile(tcc, cName, directory = None, min = 0):
if not directory:
fN = tcc + '.png'
else:
fN = directory + '/' + tcc + '.png'
tccStretch = cgPeaks.stretch(tcc, cName)
highest = tccStretch.getHighestLevel()
if highest < min:
return 0
sortedX = tccStretch.profile.keys()
sortedX.sort()
sortedY = []
for X in sortedX:
sortedY.append(tccStretch.profile[X])
gDevice = importr('grDevices')
gDevice.png(file=fN, width=1680, height=1050)
r.plot(sortedX, sortedY, xlab = "Coordinates", ylab = "Expression Level")
r.lines(sortedX, sortedY, type = "b")
gDevice.dev_off()
def main(argv=None):
"""script main.
parses command line options in sys.argv, unless *argv* is given.
"""
if argv is None:
argv = sys.argv
# setup command line parser
parser = E.OptionParser(version="%prog version: $Id$",
usage=globals()["__doc__"])
parser.add_option("-g", "--gtf-file", dest="gtf", type="string",
help="GTF containing gene annotations")
parser.add_option("-s", "--sort", dest="sort", type="choice",
default="length",
choices=sort_choices,
help="Property to sort rows by. Choices are %s"
% ", ".join(sort_choices))
parser.add_option("-b", "--bin-size", dest="bin_size", type="int",
default=25,
help="Size of window over which to sum reads")
parser.add_option("-u", "--upstream-window", dest="us_win", type="int",
default=500,
help="Amount of sequence upstream of alignment point (less introns)")
parser.add_option("-d", "--downstream-window", dest="ds_win", type="int",
default=None,
help="Amount of sequence downstream of alignment point (default longest segment)")
parser.add_option("-a", "--align-at", dest="align_at", type="choice",
default="start",
choices=align_choices,
help="Where to align genes/transcripts at. Choices are %s"
% ", ".join(align_choices))
parser.add_option("-H", "--height", dest="height", type="int",
default=None,
help="Number of rows in output matrix/heigh of plot in px")
parser.add_option("-w", "--width", dest="width", type="int",
default=None,
help="Number of columns in output/width of plot in px"
"default based on bin size")
parser.add_option("-n", "--normalize", dest="normalize", type="choice",
default="none",
choices=norm_choices,
help="Row normalization to apply. Choices are: %s"
% ", ".join(norm_choices))
parser.add_option("-r", "--renormalize", dest="renormalize", type="choice",
default="none",
choices=norm_choices,
help="Row normalization to apply after row/column compression")
parser.add_option("--no-plot", dest="plot", action="store_false",
default=True,
help="Do not output plot - compute matrix only")
parser.add_option("--use-matrix", dest="use_matrix", type="string",
default=None,
help="Use existing matrix")
parser.add_option("--annotations", dest="annotations", type="choice",
action="append",
choices=annotation_choices,
help="Add annotations to the output plot")
parser.add_option("--reverse-strand", dest="rstrand", action="store_true",
default=False,
help="Find reads on reverse strand")
parser.add_option("-f", "--feature", dest="feature", type="choice",
choices=["gene", "transcript"],
default="gene",
help="use genes or transcripts")
parser.add_option("--quantile", dest="quantile", type="float",
default=0.99,
help="Quantile to use in quantile normalization")
parser.add_option("-o", "--outfile-prefix", dest="outfile_pattern", type="string",
default=None,
help="base of names for output files")
parser.add_option("-c", "--crop", dest="crop", type="string",
default=None,
help="crop view to a certain range on the xaxis. Specify like"
"-500:1000")
parser.add_option("--format", dest="format", type="string",
default="png",
help="Output format, use valid R graphics device")
parser.add_option("--plus-wig", dest="plus_wig", type="string",
help="Use this wig for plus strand info rather than bam file")
parser.add_option("--minus-wig", dest="minus_wig", type="string",
help="Use this wig for minus strand info rather than bam file")
parser.add_option("--bed", dest="bed", type="string",
help="Use this bed for signal(must be indexed)")
parser.add_option("--norm-mat", dest="norm_mat", type="string",
help="Use this matrix for normalizing (e.g. RNA data")
parser.add_option("--sort-order-file", dest="sort_file", type="string",
default=None,
help="Two column file containing gene names in the first
column and a numeric value to sort on in the second")
# add common options (-h/--help, ...) and parse command line
(options, args) = E.Start(parser, argv=argv)
if options.plot and (options.height is None):
options.height = 100
if options.gtf:
f = IOTools.openFile(options.gtf)
if options.feature == "gene":
gtf_iterator = GTF.flat_gene_iterator(GTF.iterator(f))
else:
gtf_iterator = GTF.transcript_iterator(GTF.iterator(f))
lengths = dict()
utr3_lengths = dict()
utr5_lengths = dict()
first_exon_lengths = dict()
for transcript in gtf_iterator:
lengths[transcript[0].transcript_id] = sum(
[e[1] - e[0] for e in GTF.asRanges(transcript, "exon")])
exons = GTF.asRanges(transcript, "exon")
utrs = GTF.asRanges(transcript, "UTR")
coding = Intervals.truncate(exons, utrs)
coding.sort()
utr5 = [utr for utr in utrs if utr[1] <= coding[0][0]]
utr3 = [utr for utr in utrs if utr[0] >= coding[-1][-1]]
if transcript[0].strand == "-":
utr3, utr5 = utr5, utr3
if transcript[0].strand == "+" or len(exons) == 1:
first_exon_lengths[transcript[0].transcript_id] = \
exons[0][1] - exons[0][0]
else:
first_exon_lengths[transcript[0].transcript_id] = \
exons[-1][1] - exons[-1][0]
utr3_lengths[transcript[0].transcript_id] = sum(
[e[1] - e[0] for e in utr3])
utr5_lengths[transcript[0].transcript_id] = sum(
[e[1] - e[0] for e in utr5])
lengths = pandas.Series(lengths)
utr3_lengths = pandas.Series(utr3_lengths)
utr5_lengths = pandas.Series(utr5_lengths)
first_exon_lengths = pandas.Series(first_exon_lengths)
else:
options.sort = "none"
options.annotations = None
if options.plus_wig:
getter = iCLIP.make_getter(plus_wig=options.plus_wig,
minus_wig=options.minus_wig)
elif options.bed:
getter = iCLIP.make_getter(bedfile=options.bed)
else:
try:
getter = iCLIP.make_getter(bamfile=args[0])
except IOError:
E.error("Cannot open bamfile %s" % args[0])
return(1)
except IndexError:
getter = None
if options.use_matrix:
raw_matrix = pandas.read_csv(options.use_matrix,
sep="\t",
index_col=0)
raw_matrix.columns = raw_matrix.columns.astype("int")
else:
raw_matrix = get_matrix(getter, lengths, options)
if options.crop:
crop_from, crop_to = map(int, options.crop.split(":"))
raw_matrix = raw_matrix.loc[:, crop_from:crop_to]
if options.norm_mat:
norm_matrix = pandas.read_csv(options.norm_mat,
sep="\t",
index_col=0)
norm_matrix.columns = norm_matrix.columns.astype("int")
if options.crop:
norm_matrix = norm_matrix.loc[:, crop_from:crop_to]
if all(norm_matrix.columns == raw_matrix.columns) and \
all(raw_matrix.index.isin(norm_matrix.index.values)):
norm_matrix = norm_matrix.loc[raw_matrix.index]
norm_matrix = norm_matrix.replace(
0, norm_matrix[norm_matrix > 0].min().min())
raw_matrix = raw_matrix/norm_matrix
norm_matrix = None
else:
raise ValueError("Incompatible normalisation matrix")
normalized_matrix = normalize(raw_matrix, options.normalize,
quantile=options.quantile)
if options.sort == "length":
sorter = lengths
elif options.sort == "3utr":
sorter = utr3_lengths
elif options.sort == "5utr":
sorter = utr5_lengths
elif options.sort == "first-exon":
sorter = first_exon_lengths
elif options.sort == "manual":
sorter = pandas.read_csv(options.sort_file, sep="\t",
index_col=0, usecols=[0, 1])
sorter = sorter[sorter.columns[0]]
elif options.sort == "none":
sorter = pandas.Series(range(raw_matrix.shape[0]),
index=raw_matrix.index[::-1])
sorter = sorter[sorter.index.isin(normalized_matrix.index)]
sorter = sorter.sort_values(ascending=False)
sorted_matrix = normalized_matrix.loc[sorter.index.values]
compress_matrix = iCLIP.compress_matrix(sorted_matrix,
ncols=options.width,
nrows=options.height)
renormalized_matrix = normalize(compress_matrix, options.renormalize,
quantile=options.quantile)
if renormalized_matrix is raw_matrix and options.use_matrix is not None:
E.info("Input and output matrices are identical, no matrix output")
else:
if options.outfile_pattern:
mat_outfile = IOTools.openFile(
options.outfile_pattern + ".matrix.tsv.gz", "w")
else:
mat_outfile = options.stdout
renormalized_matrix.to_csv(mat_outfile, sep="\t")
if options.plot:
try:
from rpy2.robjects import r as R
from rpy2 import robjects as ro
except:
E.info("No rpy2. Not plotting image")
return(0)
from rpy2.robjects.numpy2ri import numpy2ri
ro.conversion.py2ri = numpy2ri
ro.numpy2ri.activate()
if options.outfile_pattern:
plot_outfile = options.outfile_pattern + ".png"
else:
plot_outfile = "bam2heatmap_out.png"
c = R["c"]
R[options.format](plot_outfile,
width=renormalized_matrix.shape[1] + 72,
height=renormalized_matrix.shape[0] + 72,
unit="px",
res=72)
R.par(mai=c(1, 0.5, 0, 0.5))
cols = R["colorRampPalette"](c("white", "blue"))(50)
bases = renormalized_matrix.columns.values.astype("int")
groups = renormalized_matrix.index.values.astype("int")
mat = renormalized_matrix.as_matrix()
mat[mat >= 1] = 1
R.image(bases, groups, R.t(mat),
zlim=c(0, 1),
raster=True,
col=cols,
xlab="Base",
yaxt="n")
def _sort_and_compress_annotation(anno):
sorted_anno = anno.loc[sorter.index]
comp_anno = iCLIP.compress_matrix(
sorted_anno, renormalized_matrix.shape[0])
return comp_anno
if options.annotations:
ends = _sort_and_compress_annotation(lengths)
starts = pandas.Series(0, index=renormalized_matrix.index)
if options.align_at == "end":
starts, ends = -1 * ends, starts
if "start" in options.annotations:
R.lines(starts.values, starts.index.values, col="black", pch=".")
if "end" in options.annotations:
R.lines(ends.values, ends.index.values,
pch=".", col="black")
if "5utr" in options.annotations:
utr5s = _sort_and_compress_annotation(utr5_lengths)
utr5s = starts + utr5s
R.lines(utr5s.values, utr5s.index.values, col="orange", pch=".")
if "3utr" in options.annotations:
utr3s = _sort_and_compress_annotation(utr3_lengths)
utr3s = ends - utr3s
R.lines(utr3s.values, utr3s.index.values, col="orange", pch=".")
R["dev.off"]()
# write footer and output benchmark information.
E.Stop()