def plot_boxplot_r(self, filename=None, n_highest_weights=1, n_highest_weights_for_quantity=1, weight_threshold=None, logy=False):
from rpy import r
logstring = ''
if logy:
logstring='y'
if filename is not None:
r.pdf(file=filename)
for var, values in self.values_from_mr.iteritems():
plot_one_boxplot_r(values, var, logstring)
if values.ndim == 1:
v = resize(values, (1, values.size))
else:
v = values
ivar = self.get_index_for_quantity(var)
if weight_threshold is not None:
for i in range(0, v.shape[0]):
iw = self.get_index_of_component_weights_over_threshold(ivar, weight_threshold)
if iw.size > 0:
r.points(i+1, v[i:(i+iw.size),iw], col='yellow', cex=0.5)
iw = self.get_index_of_weights_over_threshold(weight_threshold)
if iw.size > 0:
r.points(i+1, v[i:(i+iw.size),iw], col='blue', cex=0.5)
if n_highest_weights_for_quantity > 0:
for i in range(0, v.shape[0]):
r.points(i+1, v[i,self.get_index_of_n_highest_component_weights(ivar, n_highest_weights_for_quantity)], col='green', cex=0.5)
if n_highest_weights > 0:
for i in range(0, v.shape[0]):
r.points(i+1, v[i,self.get_index_of_n_highest_weights(n_highest_weights)], col='red', cex=0.5)
if filename is not None:
r.dev_off()
def main():
in_fname = sys.argv[1]
out_fname = sys.argv[2]
try:
columns = int( sys.argv[3] ) - 1, int( sys.argv[4] ) - 1
except:
stop_err( "Columns not specified, your query does not contain a column of numerical data." )
title = sys.argv[5]
xlab = sys.argv[6]
ylab = sys.argv[7]
matrix = []
skipped_lines = 0
first_invalid_line = 0
invalid_value = ''
invalid_column = 0
i = 0
for i, line in enumerate( file( in_fname ) ):
valid = True
line = line.rstrip( '\r\n' )
if line and not line.startswith( '#' ):
row = []
fields = line.split( "\t" )
for column in columns:
try:
val = fields[column]
if val.lower() == "na":
row.append( float( "nan" ) )
else:
row.append( float( fields[column] ) )
except:
valid = False
skipped_lines += 1
if not first_invalid_line:
first_invalid_line = i + 1
try:
invalid_value = fields[column]
except:
invalid_value = ''
invalid_column = column + 1
break
else:
valid = False
skipped_lines += 1
if not first_invalid_line:
first_invalid_line = i + 1
if valid:
matrix.append( row )
if skipped_lines < i:
try:
r.pdf( out_fname, 8, 8 )
r.plot( array( matrix ), type="p", main=title, xlab=xlab, ylab=ylab, col="blue", pch=19 )
r.dev_off()
except Exception, exc:
stop_err( "%s" % str( exc ) )
def plots(regression_o, getData_o):
"""Plots the dataset with a regression line and a boxplot using R."""
fname1 = 'car_regress.pdf'
r.pdf(fname1)
r.plot(getData_o, ylab='dist', xlab='speed')
r.abline(regression_o['(Intercept)'], regression_o['y'], col='red')
r.dev_off()
fname2 = 'car_hist.pdf'
r.pdf(fname2)
r.boxplot(getData_o, names=['dist', 'speed'])
r.dev_off()
return fname1, fname2
def output(self):
self.matrix = array(self.result_array)
p_value_list = map(str, self.p_value_list)
self.of.write('p_value\t%s\n'%'\t'.join(p_value_list))
df = self.df_lower
for cor_list in self.result_array:
cor_list = map(str, cor_list) #string can be 'join'ed. easy to output
self.of.write('%d\t%s\n'%(df, '\t'.join(cor_list)))
df += 1
r.pdf('p_value_cor.pdf')
#select a column to plot
cor_list = self.matrix[:,self.column]
p_value_label = self.p_value_list[self.column]
df_list = range(self.df_lower, self.df_upper+1)
r.plot(df_list, cor_list, type='o', pch='*', xlab='df', ylab='correlation', main='p_value: %s'%p_value_label)
r.dev_off()
def plot(self):
vertex_list = self.graph.node_list()
number_of_nodes = len(vertex_list)
for vertex in vertex_list:
degree = self.graph.inc_degree(vertex) + self.graph.out_degree(vertex)
if degree not in self.degree_dict:
self.degree_dict[degree] = 1
else:
self.degree_dict[degree] += 1
r.pdf('degree_distribution.pdf')
x_list = []
y_list = []
for degree in self.degree_dict:
x_list.append(r.log(degree))
y_list.append(r.log(float(self.degree_dict[degree])/number_of_nodes))
r.plot(x_list, y_list, type='p', xlab='log(k)', ylab='log(p(k))')
r.dev_off()
def main(args):
sourcefiles = get_src_files(args.get('in'))
hashofhos = None
parsedfiles = []
for i in range(len(sourcefiles) - 1):
queryfile = sourcefiles[i]
subjectfile = sourcefiles[i + 1]
if hashofhos:
idfile = 'keepids.tmp'
fw = open(idfile, 'w')
for id in hashofhos.keys():
fw.write(id + '\n')
fw.flush()
fw.close()
outfile = 'red_' + get_basename(queryfile) + '.aa'
os.system('reduce_fasta_file.py -f %s -i %s -o %s' %
(queryfile, idfile, outfile))
queryfile = outfile
blastout = blast(queryfile, subjectfile)
parsedfile, hashofhos = parse_blastout(blastout, args)
parsedfiles.append(parsedfile)
infomsg("hits: %s" % len(hashofhos))
Homologs = integrate_all_homologs(parsedfiles, args)
# stats
no = []
for sid, orthlist in Homologs.iteritems():
n = len(orthlist) + 1
# infomsg( str(n) )
no.append(n)
from rpy import r
outfile = 'hist_size_homol_sets.pdf'
title = 'Size of Homologous Sets'
x = 'number of homologs'
y = 'frequency'
r.pdf(outfile)
r.hist(no, xlab=x, ylab=y, main=title, col='grey', breaks=max(no))
r.dev_off()
def plot_values_as_boxplot_r(values_dict, filename=None, logy=False):
"""Create a set of boxplots (using R), one plot per variable in values_dict (dictionary of
varible name and values (1- or 2-D array)), one box per row.
If filename is given, the plot goes into that file as pdf. If 'logy' is True, the y-axis
is plotted on the log scale.
"""
from rpy import r
logstring = ''
if logy:
logstring='y'
if filename is not None:
r.pdf(file=filename)
for var, values in values_dict.iteritems():
plot_one_boxplot_r(values, var, logstring)
if filename is not None:
r.dev_off()
def main( args ):
sourcefiles = get_src_files( args.get('in') )
hashofhos = None
parsedfiles = []
for i in range( len(sourcefiles)-1 ):
queryfile = sourcefiles[i]
subjectfile = sourcefiles[i+1]
if hashofhos:
idfile = 'keepids.tmp'
fw = open( idfile, 'w' )
for id in hashofhos.keys(): fw.write( id + '\n')
fw.flush()
fw.close()
outfile = 'red_' + get_basename(queryfile) + '.aa'
os.system( 'reduce_fasta_file.py -f %s -i %s -o %s' %(queryfile,idfile,outfile) )
queryfile = outfile
blastout = blast( queryfile, subjectfile )
parsedfile, hashofhos = parse_blastout( blastout, args )
parsedfiles.append( parsedfile )
infomsg( "hits: %s" %len(hashofhos) )
Homologs = integrate_all_homologs( parsedfiles, args )
# stats
no = []
for sid, orthlist in Homologs.iteritems():
n = len(orthlist) + 1
# infomsg( str(n) )
no.append(n)
from rpy import r
outfile = 'hist_size_homol_sets.pdf'
title = 'Size of Homologous Sets'
x = 'number of homologs'
y = 'frequency'
r.pdf( outfile )
r.hist(no, xlab=x, ylab=y, main=title, col='grey', breaks=max(no))
r.dev_off()
def __init__(self, bedfile, dir, win, ma, out, ymin=0, ymax=0):
self.L = 17 # number of letters per line
Bed.__init__(self, bedfile)
self.dir = dir
self.win = win + ma
self.ma = ma # window for moving average
try:
from rpy import r
r.pdf(out + '.ConservationPlot.pdf')
self.Run()
if not ymin:
ymin = self.mscore[self.mscore > 0].min() - 0.05
if not ymax:
ymax = self.mscore.max() + 0.05
r.plot(range(-1*win, win+1), self.mscore, type = 'l', xlab = \
'Distance from the Center of Enriched Regions', \
ylab = 'Conservation Score', lwd= 3, ylim = (ymin, ymax))
r.dev_off()
except:
print >> sys.stderr, 'error import r using rpy, will not generate phastCons plot'
print sys.exc_info()[0], sys.exc_info()[1]
def plot(self):
r.pdf("per_p_value05.pdf")
r.plot(self.dataset_no, self.per_05, type='o', pch='*', xlab='dataset no.',\
ylab='percentage', main='p_value: 0.05. #edges compared with correlation cut_off 0.6')
r.dev_off()
r.pdf("per_p_value025.pdf")
r.plot(self.dataset_no, self.per_025, type='o', pch='*', xlab='dataset no.',\
ylab='percentage', main='p_value: 0.025. #edges compared with correlation cut_off 0.6')
r.dev_off()
r.pdf("per_p_value01.pdf")
r.plot(self.dataset_no, self.per_01, type='o', pch='*', xlab='dataset no.',\
ylab='percentage', main='p_value: 0.01. #edges compared with correlation cut_off 0.6')
r.dev_off()
def plot(self, filename, list_to_plot, main_lab, xlab):
max_length = max(list_to_plot)
r.pdf(filename)
r.hist(list_to_plot, breaks=range(max_length + 1), las=1, main=main_lab, xlab=xlab)
r.dev_off()
# Make the CanonicalProperties
try:
cp = CanonicalProperties(options.muninn_log_file, options.which)
except CanonicalException, e:
print parser.error(e)
# Store all the plotting data
data = []
# Print which is used
print "Using:", cp.fullname
# Plot the required output
r.pdf(options.output, width=options.width, height=options.height)
r.par(cex=options.cex)
inv_beta = arange(options.inv_beta_min, options.inv_beta_max, 0.01)
beta = 1.0/inv_beta
lnZ = vectorize(cp.lnZ)(beta)
r.plot(inv_beta, lnZ, type='l', xlab=r("expression(beta**-1)"), ylab=r("""expression(paste("ln ", Z(beta)))"""))
data.append((cp.number, "lnZ", (inv_beta, lnZ)))
betaF = vectorize(cp.betaF)(beta)
r.plot(inv_beta, betaF, type='l', xlab=r("expression(beta**-1)"), ylab=r("expression(F(beta) * beta)"))
data.append((cp.number, "betaF", (inv_beta, betaF)))
S = vectorize(cp.S)(beta)
r.plot(inv_beta, S, type='l', xlab=r("expression(beta**-1)"), ylab=r("expression(S(beta) / k[B])"))
def plot(outfile, data, out_format='png'):
w = int(round(len(data)/4.0))
if out_format == 'png':
r.png(outfile, width=w*100, height=1000, res=72)
elif out_format == 'pdf':
r.pdf(outfile, width=w, height=10)
else:
raise Exception('Unrecognised format: ' + str(out_format))
print("total: " + str(len(data)))
series = []
points = {'translate': [], 'preprocessing': []}
for dat in data:
points['translate'].append(float(dat['translate']))
points['preprocessing'].append(float(dat['preprocessing']))
xlabels = []
for k, v in data[0].iteritems():
if k not in ["problem", 'translate', 'preprocessing']:
series.append(k)
points[k] = []
index = 0
for dat in data:
for k in series:
if dat[k] != 'no-plan':
points[k].append(float(dat[k]) + \
points['translate'][index] + \
points['preprocessing'][index])
else:
points[k].append(-1000)
xlabels.append(dat['problem'])
index += 1
max_value = max(iter([max(iter(points[k])) for k in series]))
yrange = (0, max_value)
legend_labels = []
x = [i for i in range(1,len(points['translate'])+1)]
y = [-1000 for i in x]
r.par(mar=(7,5,4,2))
r.plot(x, y, main='', xlab="", ylab='',
xaxt='n', yaxt='n', pch=0, ylim=yrange,
mgp=(5,1,0))
r.mtext("Problem", side=1, line=5)
r.mtext("CPU Time (s)", side=2, line=3)
pch_start = 1
pch_index = pch_start
# plotting "translate"
#r.plot(x, points['translate'], main='',
# xlab='', ylab='Time (s)',
# xaxt='n', yaxt='n',
# pch=0, ylim=yrange)
#legend_labels.append('translate')
r.lines(x, points['translate'], lty=1)
# preprocessing -- Removed since it's insignificant
#r.points(x, points['preprocessing'], pch=pch_index)
#pch_index =+ 1
# planner output
for k in series:
if k != 'translate' and k != 'preporcessing':
r.points(x, points[k], pch=pch_index)
pch_index += 1
legend_labels.append("FD+" + k.upper())
# put x-axis labels
for i in range(0, len(xlabels)):
r.axis(side=1, at=i+1, labels=xlabels[i], las=2)
# put y-axis labels
base, step = get_y_step(max_value)
print("base: " + str(base) + " -- step: " + str(step))
y = base
for i in range(0, step):
r.axis(side=2, at=y, labels=str(y), las=2)
y += base
# legend
r.legend(1, max_value, legend_labels, pch=[i for i in range(pch_start, pch_index)])
r.dev_off()
type='float',
help="ymin in the plot")
parser.add_option("--ymax",
dest="ymax",
default=0,
type='float',
help="ymax in the plot")
(options, args) = parser.parse_args()
if len(sys.argv) < 2:
parser.print_help()
sys.exit()
if options.make:
MakeBinary(sys.argv[1])
sys.exit()
for bed in args:
out = bed.split('.')[0]
phastCons(bed, options.dir, options.range, options.win, out,
options.ymin, options.ymax)
'''
from rpy import r
self = phastCons('bedfile', '/misc/iris/acct/weili/database/humanhg17_May2004/phastCons', 3000, 250)
r.pdf('ConservationPlot.pdf')
r.plot(range(-3000, 3001), self.mscore, type = 'l', xlab = 'Distance from Center of Binding Sites', ylab = 'Conservation Score', lwd= 4, xlim = (-2000, 2000), ylim = (0.10, 0.25))
r.dev_off()
'''
def plot(outfile, data, out_format='png'):
w = int(round(len(data) / 4.0))
if out_format == 'png':
r.png(outfile, width=w * 100, height=1000, res=72)
elif out_format == 'pdf':
r.pdf(outfile, width=w, height=10)
else:
raise Exception('Unrecognised format: ' + str(out_format))
print("total: " + str(len(data)))
series = []
points = {'translate': [], 'preprocessing': []}
for dat in data:
points['translate'].append(float(dat['translate']))
points['preprocessing'].append(float(dat['preprocessing']))
xlabels = []
for k, v in data[0].iteritems():
if k not in ["problem", 'translate', 'preprocessing']:
series.append(k)
points[k] = []
index = 0
for dat in data:
for k in series:
if dat[k] != 'no-plan':
points[k].append(float(dat[k]) + \
points['translate'][index] + \
points['preprocessing'][index])
else:
points[k].append(-1000)
xlabels.append(dat['problem'])
index += 1
max_value = max(iter([max(iter(points[k])) for k in series]))
yrange = (0, max_value)
legend_labels = []
x = [i for i in range(1, len(points['translate']) + 1)]
y = [-1000 for i in x]
r.par(mar=(7, 5, 4, 2))
r.plot(x,
y,
main='',
xlab="",
ylab='',
xaxt='n',
yaxt='n',
pch=0,
ylim=yrange,
mgp=(5, 1, 0))
r.mtext("Problem", side=1, line=5)
r.mtext("CPU Time (s)", side=2, line=3)
pch_start = 1
pch_index = pch_start
# plotting "translate"
#r.plot(x, points['translate'], main='',
# xlab='', ylab='Time (s)',
# xaxt='n', yaxt='n',
# pch=0, ylim=yrange)
#legend_labels.append('translate')
r.lines(x, points['translate'], lty=1)
# preprocessing -- Removed since it's insignificant
#r.points(x, points['preprocessing'], pch=pch_index)
#pch_index =+ 1
# planner output
for k in series:
if k != 'translate' and k != 'preporcessing':
r.points(x, points[k], pch=pch_index)
pch_index += 1
legend_labels.append("FD+" + k.upper())
# put x-axis labels
for i in range(0, len(xlabels)):
r.axis(side=1, at=i + 1, labels=xlabels[i], las=2)
# put y-axis labels
base, step = get_y_step(max_value)
print("base: " + str(base) + " -- step: " + str(step))
y = base
for i in range(0, step):
r.axis(side=2, at=y, labels=str(y), las=2)
y += base
# legend
r.legend(1,
max_value,
legend_labels,
pch=[i for i in range(pch_start, pch_index)])
r.dev_off()
