Mercurial > repos > matces > carpet_toolsuite

comparison carpet-src-1/tools/CARPET/TSS_distance.py @ 0:cdd489d98766

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Migrated tool version 1.0.0 from old tool shed archive to new tool shed repository
author matces
date Tue, 07 Jun 2011 16:50:41 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:cdd489d98766
1 #!/usr/bin/env python
2
3 # Copyright 2009 Matteo Cesaroni, Lucilla Luzi
4 #
5 # This program is free software; ; you can redistribute it and/or modify
6 # it under the terms of the GNU Lesser General Public License as published by
7 # the Free Software Foundation; either version 3 of the License, or (at your
8 # option) any later version.
9 #
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
14
15
16 import sys
17 from rpy import *
18
19
20 def stop_err(msg):
21 sys.stderr.write(msg)
22 sys.exit()
23
24 def main():
25
26 # Handle input params
27 in_fname = sys.argv[1]
28 out_fname = sys.argv[2]
29 try:
30 column = int( sys.argv[3] ) - 1
31 except:
32 stop_err( "..Column not specified, your query does not contain a column of numerical data." )
33 title = sys.argv[4]
34 xlab = sys.argv[5]
35 breaks = int( sys.argv[6] )
36 if breaks == 0: breaks = "Sturges"
37 if sys.argv[7] == "true": density = True
38 else: density = False
39
40
41
42 matrix = []
43 skipped_lines = 0
44 first_invalid_line = 0
45 invalid_value = ''
46
47 for i, line in enumerate( file( in_fname ) ):
48 valid = True
49 line = line.rstrip('\r\n')
50 # Skip comments
51 if line and not line.startswith( '#' ):
52 # Extract values and convert to floats
53 row = []
54 try:
55 fields = line.split( "\t" )
56 val = fields[column]
57 if val.lower() == "na":
58 row.append( float( "nan" ) )
59 if float(val) > float(xlab):
60 val = (float(xlab)+2000)
61
62 row.append( float( val ) )
63 except:
64 valid = False
65 skipped_lines += 1
66 if not first_invalid_line:
67 first_invalid_line = i+1
68 else:
69 try:
70 row.append( float( val ) )
71 except ValueError:
72 valid = False
73 skipped_lines += 1
74 if not first_invalid_line:
75 first_invalid_line = i+1
76 invalid_value = fields[column]
77 else:
78 valid = False
79 skipped_lines += 1
80 if not first_invalid_line:
81 first_invalid_line = i+1
82
83 if valid:
84 matrix.append( row )
85
86 if skipped_lines < i:
87 print "..on columnn %s" %sys.argv[3]
88 try:
89 a = array( matrix )
90 r.pdf( out_fname, 8, 8 )
91 r.hist( a, probability=True, main=title, xlab="TSS distance", breaks=breaks )
92 if density:
93 r.lines( r.density( a ) )
94 r.dev_off()
95 except exc:
96 stop_err("Building histogram resulted in error: %s." %str( exc ))
97 else:
98 print "..all values in column %s are non-numeric." %sys.argv[3]
99
100 if skipped_lines > 0:
101 print "..skipped %d invalid lines starting with line #%d. Value '%s' is not numeric." % ( skipped_lines, first_invalid_line, invalid_value )
102
103 r.quit( save="no" )
104
105 if __name__ == "__main__":
106 main()