cpgtools 2.0.5__py3-none-any.whl

cpgtools-2.0.5.data/scripts/beta_combat.py

@@ -0,0 +1,174 @@
+#!python
+
+"""
+Description
+-----------
+This program corrects batch effect using the "combat" algorithm:
+
+W. Evan Johnson, et al, Adjusting batch effects in microarray expression data using empirical Bayes methods, Biostatistics, 2007.
+
+Example of input data file
+---------------------------
+CpG_ID    Sample_01    Sample_02    Sample_03    Sample_04
+cg_001    0.831035    0.878022    0.794427    0.880911
+cg_002    0.249544    0.209949    0.234294    0.236680
+cg_003    0.845065    0.843957    0.840184    0.824286
+...
+
+Example of batch file
+-------------------------------
+Sample,Group
+Sample_01,plate_1
+Sample_02,plate_1
+Sample_03,plate_2
+Sample_04,plate_2
+...
+
+Notes
+-----
+* Rows with missing values will be removed
+
+"""
+
+
+import sys
+import subprocess
+from optparse import OptionParser
+from cpgmodule.utils import *
+from cpgmodule._version import __version__
+import pandas as pd
+#from sklearn.preprocessing import StandardScaler
+#from sklearn.decomposition import PCA
+from sklearn.impute import KNNImputer
+from collections import Counter
+from combat.pycombat import pycombat
+import matplotlib.pyplot as plt
+import matplotlib.colors as mcolors
+from impyutelib import nan_indices
+
+
+__author__ = "Liguo Wang"
+__copyright__ = "Copyleft"
+__credits__ = []
+__license__ = "GPL"
+__maintainer__ = "Liguo Wang"
+__email__ = "wang.liguo@mayo.edu"
+__status__ = "Development"
+
+def pick_colors(n):
+    my_colors = list(mcolors.CSS4_COLORS.keys())
+    if n > len(my_colors):
+        print ("Too many colors requested", file = sys.stderr)
+        sys.exit()
+    return my_colors[0:n]
+
+def box_plot(df, s_colors,  out_png, ylab="Beta values", title=""):
+    s_names = df.columns
+    fig, ax = plt.subplots()
+    bplot = ax.boxplot(df, patch_artist=True, tick_labels = s_names)
+    for patch, color in zip(bplot['boxes'], s_colors):
+        patch.set_facecolor(color)
+    ax.set_xticklabels(s_names, rotation='vertical')
+    plt.ylabel(ylab)
+    plt.title(title)
+    plt.savefig(out_png)
+
+def main():
+    
+    usage="%prog [options]" + "\n"
+    parser = OptionParser(usage,version="%prog " + __version__)
+    parser.add_option("-i","--input_file",action="store",type="string",dest="input_file",help="Tab-separated data frame file containing beta values with the 1st row containing sample IDs and the 1st column containing CpG IDs.")
+    parser.add_option("-k",action="store",type="int", default=3, dest="n_neighbors",help="Number of neighbors to use for imputation. default=%default")
+    parser.add_option("--axis",type="choice",choices=[0,1],default=1,dest="axis_choice",help="How to do imputation (using the KNN algorithm) if the input file has missing values. 1: search columns for k nearest neighbours; 0: Search rows for k nearest neighbours. default=%default")
+    parser.add_option("-g","--group",action="store",type="string",dest="group_file",help="Comma-separated group file defining the batch groups of each sample.")
+    parser.add_option("-o","--output",action="store",type='string', dest="out_file",help="The prefix of the output file.")
+    (options,args)=parser.parse_args()
+    
+    print ()
+    if not (options.input_file):
+        print (__doc__)
+        parser.print_help()
+        sys.exit(101)
+    if not (options.group_file):
+        print (__doc__)
+        parser.print_help()
+        sys.exit(101)    
+    if not (options.out_file):
+        print (__doc__)
+        parser.print_help()
+        sys.exit(103)    
+
+    beta_out1 = options.out_file + '.combat.tsv'
+    beta_out2 = options.out_file + '.combat_withNAs.tsv'
+    beta_boxplot_before = options.out_file + '.boxplot.png'
+    beta_boxplot_after = options.out_file + '.boxplot_combat.png'
+    
+    printlog("Reading input file: \"%s\" ..." % (options.input_file))
+    df1 = pd.read_csv(options.input_file, index_col = 0, sep="\t")
+    
+    # count NA
+    total_na = df1.isna().sum().sum()
+    printlog("The input file: \"%s\" contains %d missing values" % (options.input_file, total_na))
+
+    # if there are any NAs, either impute or remove
+    if total_na > 0:
+        printlog("Imputing missing values using KNN ...")
+        na_locations = nan_indices(df1.to_numpy())
+        imputer = KNNImputer(n_neighbors = options.n_neighbors)
+        if options.axis_choice == 1:
+            input_df = df1.T
+            after = imputer.fit_transform(input_df)
+            df2 = pd.DataFrame(after, index = input_df.index,
+                                  columns = input_df.columns).T
+            #output_df = output_df.round(args.decimal)
+        elif options.axis_choice == 0:
+            input_df = df1
+            after = imputer.fit_transform(input_df)
+            df2 = pd.DataFrame(after, index = input_df.index,
+                                  columns = input_df.columns)
+        else:
+            raise ValueError("axis only accepts 0 or 1.")
+    else:
+        df2 = df1
+
+    printlog("Reading group file: \"%s\" ..." % (options.group_file))
+    group = pd.read_csv(options.group_file, index_col=0, names=['Sample_ID', 'Group_ID'])
+    #check if sample IDs are unique
+    if len(group.index) != len(group.index.unique()):
+        print ("Sample IDs are not unique", file = sys.stderr)
+        sys.exit()    
+    group.index = group.index.map(str)
+    printlog("Group/batch \"%s\" contains %d samples" % (options.group_file, len(group.index)))
+    
+    # a list of unique group names, and their frequencies
+    group_info = Counter(group['Group_ID'])
+    print(list(group['Group_ID']))
+    # a list of unique colors
+    color_names = pick_colors(len(group_info))    
+    color_list = []
+    for name,count in zip(color_names, list(group_info.values())):
+        color_list.extend([name]*count)
+
+    printlog("Generate boxplot before correction. Save to '%s'" % beta_boxplot_before)
+    box_plot(df2, s_colors=color_list, out_png=beta_boxplot_before, title="Before batch effects correction")
+
+    # remove batch effect
+    printlog("Removing batch effect ...")
+    df_corrected = pycombat(df2, list(group['Group_ID']))
+    
+    printlog("Generate boxplot after correction. Save to '%s'" % beta_boxplot_after)
+    box_plot(df_corrected, s_colors=color_list, out_png=beta_boxplot_after, title="After batch effects correction")
+
+    printlog("Save data after combat to \"%s\", keep imputed values ..." % beta_out1)
+    df_corrected.to_csv(beta_out1,sep="\t")
+    
+    # add the original NAs (if any) back.
+    # Only perform combat, no KNN imputation
+    if total_na > 0:
+        for i,j in na_locations:
+            df_corrected.iat[i, j] = np.nan
+        printlog("Save data after combat to \"%s\", keep orignal missing values ..." % beta_out2)
+        df_corrected.to_csv(beta_out2,sep="\t",na_rep='NA')
+
+if __name__=='__main__':
+    main()    

cpgtools-2.0.5.data/scripts/beta_jitter_plot.py

@@ -0,0 +1,107 @@
+#!python
+
+"""
+Description
+-----------
+This program generates jitter plot (a.k.a. strip chart) and bean plot for each sample (column). 
+
+Notes
+-----
+User must install the "beanplot" R library:
+https://cran.r-project.org/web/packages/beanplot/index.html
+
+Example of input
+-----------------
+CpG_ID	Sample_01	Sample_02	Sample_03	Sample_04
+cg_001	0.831035	0.878022	0.794427	0.880911
+cg_002	0.249544	0.209949	0.234294	0.236680
+cg_003	0.845065	0.843957	0.840184	0.824286
+
+Note: Please name your sample IDs using only "letters" [a-z, A-Z], "numbers" [0-9], and "_"; and
+your sample ID should start with a letter. 
+"""
+
+
+import sys,os
+import collections
+import subprocess
+import numpy as np
+from cpgmodule._version import __version__
+from optparse import OptionParser
+from cpgmodule import ireader
+from cpgmodule.utils import *
+from cpgmodule import BED
+import pandas as pd
+
+__author__ = "Liguo Wang"
+__copyright__ = "Copyleft"
+__credits__ = []
+__license__ = "GPL"
+__maintainer__ = "Liguo Wang"
+__email__ = "wang.liguo@mayo.edu"
+__status__ = "Development"
+
+	
+def main():
+	
+	usage="%prog [options]" + "\n"
+	parser = OptionParser(usage,version="%prog " + __version__)
+	parser.add_option("-i","--input_file",action="store",type="string",dest="input_file",help="Tab-separated data frame file containing beta values with the 1st row containing sample IDs and the 1st column containing CpG IDs.")
+	parser.add_option("-f","--fraction",action="store",type='float', dest="fraction", default=0.5, help="The fraction of total data points (CpGs) used to generate jitter plot. Decrease this number if the jitter plot is over-crowded. default=%default" )
+	parser.add_option("--width",action="store",type='int', default=800, dest="png_width",help="The width of the output PNG file. default=%default")
+	parser.add_option("--height",action="store",type='int', default=480, dest="png_height",help="The height of the output PNG file. . default=%default")
+	parser.add_option("-o","--output",action="store",type='string', dest="out_file",help="The prefix of the output file.")
+	(options,args)=parser.parse_args()
+	
+	print ()
+	if not (options.input_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(101)
+	
+	if not (options.out_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(103)
+	if options.fraction < 0:
+		options.fraction = 0.0
+	if options.fraction > 1:
+		options.fraction = 1.0	
+	
+	ROUT = open(options.out_file + '.r','w')
+	print ('library(beanplot)', file = ROUT)
+	print ('png(file=\"%s\", width=800, height=480, unit=\"px\")' % (options.out_file + '.png'),file=ROUT)
+	
+	
+	
+	printlog("Reading beta file: \"%s\"" % (options.input_file))
+	df = pd.read_table(options.input_file)	
+	samples = df.columns[1:]
+	if options.fraction < 1.0:
+		subset_file = options.out_file + '.sample.tsv'
+		printlog("Sampling subset (%.2f%%) from file: \"%s\"" % (options.fraction*100, options.input_file))
+		subset = df.sample(frac=options.fraction, replace=False, random_state=999)
+		printlog("Saving subset (%.2f%%) to file: \"%s\"" % (options.fraction*100, options.out_file + '.sample.tsv'))
+		subset.to_csv(options.out_file + '.sample.tsv', sep = "\t", index = False)
+		
+		print ('d = read.table(file="%s", sep="\\t", header=TRUE)' % (options.out_file + '.sample.tsv'), file=ROUT)
+		print ('ll = list(%s)' % (','.join(['"' + s + '"' + ' = d$' + s for s in samples])), file=ROUT)
+		print ('stripchart(ll,cex=0.1,col="#abd9e9", vertical=T, method=c("jitter"), ylab="Beta value",las=2, jitter=0.3,cex.names = 0.8)', file=ROUT)
+		print ('beanplot(ll, cutmin=0,cutmax=1, border="#d01c8b",what=c(1,1,1,0),col=c(), las=2, add = TRUE)', file=ROUT)
+	else:
+		printlog("Using all data points in file: \"%s\"" % (options.input_file))
+		print ('d = read.table(file="%s", sep="\\t", header=TRUE)' % (options.input_file), file=ROUT)
+		print ('ll = list(%s)' % (','.join(['"' + s + '"' + ' = d$' + s for s in samples])), file=ROUT)
+		print ('stripchart(ll,cex=0.1,col="#abd9e9", vertical=T, method=c("jitter"), ylab="Beta value", las=2, jitter=0.3,cex.names = 0.8)', file=ROUT)
+		print ('beanplot(ll, cutmin=0,cutmax=1, border="#d01c8b",what=c(1,1,1,0),col=c(), las=2, add = TRUE)', file=ROUT)
+	ROUT.close()
+	
+	try:
+		subprocess.call("Rscript " + options.out_file + '.r', shell=True)
+	except:
+		print ("Cannot generate pdf file from " + options.out_file + '.r', file=sys.stderr)
+		pass        	
+	
+if __name__=='__main__':
+	main()	
+	

cpgtools-2.0.5.data/scripts/beta_m_conversion.py

@@ -0,0 +1,105 @@
+#!python
+
+"""
+Description
+-----------
+Convert Beta-value into M-value or vice versa
+
+Example of input
+-----------------
+CpG_ID	Sample_01	Sample_02	Sample_03	Sample_04
+cg_001	0.831035	0.878022	0.794427	0.880911
+cg_002	0.249544	0.209949	0.234294	0.236680
+cg_003	0.845065	0.843957	0.840184	0.824286
+"""
+
+
+import sys,os
+import collections
+import numpy as np
+from scipy import stats
+from cpgmodule._version import __version__
+from optparse import OptionParser
+from cpgmodule import ireader
+from cpgmodule.utils import *
+
+__author__ = "Liguo Wang"
+__copyright__ = "Copyleft"
+__credits__ = []
+__license__ = "GPL"
+__maintainer__ = "Liguo Wang"
+__email__ = "wang.liguo@mayo.edu"
+__status__ = "Development"
+
+	
+def main():
+	usage="%prog [options]"
+	parser = OptionParser(usage,version="%prog " + __version__)
+	parser.add_option("-i","--input_file",action="store",type="string",dest="input_file",help="Tab-separated data frame file containing beta or M values with the 1st row containing sample IDs and the 1st column containing CpG IDs. This file can be a regular text file or compressed file (.gz, .bz2).")
+	parser.add_option("-d","--dtype",action="store",type='string', dest="data_type",default="Beta", help="Input data type either \"Beta\" or \"M\". default=%default")
+	parser.add_option("-o","--output",action="store",type='string', dest="out_file",help="The output file.")
+	(options,args)=parser.parse_args()
+	
+	print ()
+
+	if not (options.input_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(101)
+	if not (options.data_type):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(101)				
+	if not (options.out_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(103)	
+	
+	FOUT = open(options.out_file, 'w')		
+	
+	if options.data_type.lower() == "beta":
+		printlog("Convert Beta-value file \"%s\" into M-value file \"%s\" ..." % (options.input_file, options.out_file))
+	elif options.data_type.lower() == "m":
+		printlog("Convert M-value file \"%s\" into Beta-value file \"%s\" ..." % (options.input_file, options.out_file))
+	else:
+		print ("Data type must be \"Beta\" or \"M\"", file=sys.stderr)
+		sys.exit(0)
+		
+	line_num = 1
+	for l in ireader.reader(options.input_file):
+		f = l.split()
+		if line_num == 1:
+			print (l, file=FOUT)
+		else:
+			probe_ID = f[0]
+			input_values = f[1:]
+			output_values = []
+			for iv in input_values:
+				#deal with non-numerical values
+				try:
+					if options.data_type.lower() == "beta":
+						ov = np.log2(float(iv)/(1.0 - float(iv)))
+					elif options.data_type.lower() == "m":
+						ov = (2**float(iv))/(2**float(iv) + 1)
+				except:
+					ov = np.nan
+				output_values.append(ov)
+			print (probe_ID + '\t' + '\t'.join([str(i) for i in output_values]), file=FOUT)
+		line_num += 1
+
+	FOUT.close()
+	
+if __name__=='__main__':
+	main()				
+					
+					
+					
+					
+					
+					
+					
+					
+					
+					
+					
+					

cpgtools-2.0.5.data/scripts/beta_profile_gene_centered.py

@@ -0,0 +1,165 @@
+#!python
+
+"""
+Description
+-----------
+This program calculates the methylation profile (i.e. average beta value) for genomic regions 
+around genes. These genomic regions include: "5'UTR exon", "CDS exon", "3'UTR exon",
+"first intron", "internal intron", "last intron", "up-stream intergenic", and
+"down-stream intergenic".
+
+Example of input BED6+ file
+---------------------------
+chr22   44021512        44021513        cg24055475      0.9231  -
+chr13   111568382       111568383       cg06540715      0.1071  +
+chr20   44033594        44033595        cg21482942      0.6122  -
+"""
+
+
+import sys,os
+import collections
+import subprocess
+import numpy as np
+from optparse import OptionParser
+from cpgmodule._version import __version__
+from cpgmodule import ireader
+from cpgmodule.utils import *
+from cpgmodule import BED
+
+__author__ = "Liguo Wang"
+__copyright__ = "Copyleft"
+__credits__ = []
+__license__ = "GPL"
+__maintainer__ = "Liguo Wang"
+__email__ = "wang.liguo@mayo.edu"
+__status__ = "Development"
+
+def main():
+	
+	usage="%prog [options]" + "\n"
+	parser = OptionParser(usage,version="%prog " + __version__)
+	parser.add_option("-i","--input_file",action="store",type="string",dest="input_file",help="BED6+ file specifying the C position. This BED file should have at least 6 columns (Chrom, ChromStart, ChromeEnd, Name, Beta_value, Strand). BED6+ file can be a regular text file or compressed file (.gz, .bz2).")
+	parser.add_option("-r","--refgene",action="store",type="string",dest="gene_file",help="Reference gene model in standard BED12 format (https://genome.ucsc.edu/FAQ/FAQformat.html#format1). \"Strand\" column must exist in order to decide 5' and 3' UTRs, up- and down-stream intergenic regions.")
+	parser.add_option("-d","--downstream",action="store",type="int",dest="downstream_size",default=2000,help="Size of down-stream genomic region added to gene. default=%default (bp)")
+	parser.add_option("-u","--upstream",action="store",type="int",dest="upstream_size",default=2000,help="Size of up-stream genomic region added to gene. default=%default (bp)")
+	parser.add_option("-o","--output",action="store",type='string', dest="out_file",help="The prefix of the output file.")
+	(options,args)=parser.parse_args()
+	
+	print ()
+	if not (options.input_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(101)
+
+	if not (options.gene_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(102)
+				
+	if not (options.out_file):
+		print (__doc__)
+		parser.print_help()
+		sys.exit(103)	
+	
+	FOUT = open(options.out_file + '.txt','w')
+	ROUT = open(options.out_file + '.r','w')
+	print ("\t".join(["Group","Relative_position(5'->3')", "Average_beta"]), file=FOUT)
+	
+	#step1: read CpG file
+	printlog("Reading CpG file: \"%s\"" % (options.input_file))
+	cpg_ranges = read_CpG_bed(options.input_file)
+		
+	#step2: read gene file
+	printlog("Reading reference gene model: \"%s\"" % (options.gene_file))
+	ref_gene = BED.ParseBED(options.gene_file)
+	
+	group_sizes = []	#number of datapoints in each group
+	printlog("Process upstream regions ...")
+	up_2k = ref_gene.getIntergenic(direction = 'up',  size=options.upstream_size)
+	s = coverage_over_range(up_2k,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Upstream_intergenic',str(i), str(s[i])]), file=FOUT)	
+	print ('Upstream_intergenic_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+	
+
+	printlog("Process 5' UTR exons ...")
+	utr5_exons = ref_gene.getUTRs(utr=5)
+	s = coverage_over_range(utr5_exons,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Five_prime_UTR',str(i), str(s[i])]), file=FOUT)	
+	print ('Five_prime_UTR_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+
+	
+	printlog("Process Coding exons ...")
+	cds_exons = ref_gene.getCDSExons()
+	s = coverage_over_range(cds_exons,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Coding_exon',str(i), str(s[i])]), file=FOUT)
+	print ('Coding_exon_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+
+	printlog("Process first introns ...")
+	introns = ref_gene.getIntrons(itype='first')
+	s = coverage_over_range(introns,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['First_intron',str(i), str(s[i])]), file=FOUT)
+	print ('First_intron_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+			
+	printlog("Process internal introns ...")
+	introns = ref_gene.getIntrons(itype='internal')
+	s = coverage_over_range(introns,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Internal_intron',str(i), str(s[i])]), file=FOUT)
+	print ('Internal_intron_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+
+
+	printlog("Process last introns ...")
+	introns = ref_gene.getIntrons(itype='last')
+	s = coverage_over_range(introns,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Last_intron',str(i), str(s[i])]), file=FOUT)
+	print ('Last_intron_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+	
+					
+	printlog("Process 3' UTR exons ...")
+	utr3_exons = ref_gene.getUTRs(utr=3)
+	s = coverage_over_range(utr3_exons,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Three_prime_UTR',str(i), str(s[i])]), file=FOUT)
+	print ('Three_prime_UTR_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+
+	printlog("Process downstream regions ...")
+	down_2k = ref_gene.getIntergenic(direction = 'down', size=options.downstream_size)
+	s = coverage_over_range(down_2k,cpg_ranges)
+	group_sizes.append(len(s))
+	for i in sorted(s):
+		print ('\t'.join(['Downstream_intergenic',str(i), str(s[i])]), file=FOUT)
+	print ('Downstream_intergenic_y=c(%s)' % ','.join([str(s[i]) for i in sorted(s)]), file=ROUT)
+	
+	print('\n')
+	print ('pdf(file=\"%s\", width=10, height=5)' % (options.out_file + '.pdf'),file=ROUT)
+	print ('plot(1:%d, c(Upstream_intergenic_y, Five_prime_UTR_y, Coding_exon_y, First_intron_y, Internal_intron_y, Last_intron_y, Three_prime_UTR_y, Downstream_intergenic_y),ylim=c(0,1), xaxt="n",xlab="", ylab="Average methylation", type="l", col="red")' % sum(group_sizes), file=ROUT)
+	print ('abline(v = c(100,201,302,403,504,605,706),col="blue", lty="dashed")', file=ROUT)
+	print ('abline(v = c(%d,%d,%d,%d,%d,%d,%d),col="blue", lty="dashed")' % (sum(group_sizes[0:1]),sum(group_sizes[0:2]),sum(group_sizes[0:3]),sum(group_sizes[0:4]),sum(group_sizes[0:5]),sum(group_sizes[0:6]),sum(group_sizes[0:7])), file=ROUT)
+	print ('abline(h = 0.5,col="grey", lty="dashed")', file=ROUT)
+	print ('text(x=c(%d,%d,%d,%d,%d,%d,%d, %d)+50, y=0.9, cex=0.7, labels=c("Upstream\\n(5\'->3\')", "5\'UTR exon\\n(5\'->3\')","Coding exon\\n(5\'->3\')","First intron\\n(5\'->3\')","Internal intron\\n(5\'->3\')","Last intron\\n(5\'->3\')", "3\'UTR exon\\n(5\'->3\')","Downstream\n(5\'->3\')"))' % (0, sum(group_sizes[0:1]),sum(group_sizes[0:2]),sum(group_sizes[0:3]),sum(group_sizes[0:4]),sum(group_sizes[0:5]),sum(group_sizes[0:6]),sum(group_sizes[0:7])), file=ROUT)
+	print ('dev.off()',file=ROUT)
+		
+	FOUT.close()
+	ROUT.close()
+	try:
+		subprocess.call("Rscript " + options.out_file + '.r', shell=True)
+	except:
+		print ("Cannot generate pdf file from " + options.out_file + '.r', file=sys.stderr)
+		pass        
+
+
+if __name__=='__main__':
+	main()	
+