line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Statistics::Reproducibility; |
2
|
|
|
|
|
|
|
|
3
|
1
|
|
|
1
|
|
40041
|
use 5.006; |
|
1
|
|
|
|
|
4
|
|
|
1
|
|
|
|
|
245
|
|
4
|
1
|
|
|
1
|
|
6
|
use strict; |
|
1
|
|
|
|
|
3
|
|
|
1
|
|
|
|
|
43
|
|
5
|
1
|
|
|
1
|
|
6
|
use warnings FATAL => 'all'; |
|
1
|
|
|
|
|
14
|
|
|
1
|
|
|
|
|
56
|
|
6
|
1
|
|
|
1
|
|
5
|
use Carp; |
|
1
|
|
|
|
|
3
|
|
|
1
|
|
|
|
|
1697
|
|
7
|
|
|
|
|
|
|
|
8
|
|
|
|
|
|
|
#use Math::Geometry::Multidimensional qw/distanceToLineN diagonalComponentsN diagonalDistancesFromOriginN/; |
9
|
1
|
|
|
1
|
|
1637
|
use Statistics::TheilSenEstimator qw/theilsen/; |
|
1
|
|
|
|
|
3869
|
|
|
1
|
|
|
|
|
67
|
|
10
|
1
|
|
|
1
|
|
8
|
use Statistics::QuickMedian qw/qmedian/; |
|
1
|
|
|
|
|
3
|
|
|
1
|
|
|
|
|
36
|
|
11
|
1
|
|
|
1
|
|
1122
|
use Statistics::Distributions; |
|
1
|
|
|
|
|
5815
|
|
|
1
|
|
|
|
|
7738
|
|
12
|
|
|
|
|
|
|
|
13
|
|
|
|
|
|
|
=head1 NAME |
14
|
|
|
|
|
|
|
|
15
|
|
|
|
|
|
|
Statistics::Reproducibility - Reproducibility measurement between multiple replicate experiments |
16
|
|
|
|
|
|
|
|
17
|
|
|
|
|
|
|
=head1 VERSION |
18
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
Version 0.09 |
20
|
|
|
|
|
|
|
|
21
|
|
|
|
|
|
|
=cut |
22
|
|
|
|
|
|
|
|
23
|
|
|
|
|
|
|
our $VERSION = '0.09'; |
24
|
|
|
|
|
|
|
|
25
|
|
|
|
|
|
|
|
26
|
|
|
|
|
|
|
=head1 SYNOPSIS |
27
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
This module facilitates investigation of reproducbility between multiple replicates of |
29
|
|
|
|
|
|
|
quantitative experiments e.g. SILAC or microarray. Scatter plots are great, but |
30
|
|
|
|
|
|
|
only 2d. Some people use correlation as a proxy for reproducibility, but it's not right. |
31
|
|
|
|
|
|
|
This module helps you through the following items... |
32
|
|
|
|
|
|
|
|
33
|
|
|
|
|
|
|
1) Summarize reproducibility across the replicates |
34
|
|
|
|
|
|
|
2) Pick out replicates that agree more or less |
35
|
|
|
|
|
|
|
3) Summarize reproducibility for individual proteins/genes/whatever |
36
|
|
|
|
|
|
|
4) Set a cutoff for what you can call significant, based on precision |
37
|
|
|
|
|
|
|
5) Deal with missing values (common in SILAC) |
38
|
|
|
|
|
|
|
|
39
|
|
|
|
|
|
|
This works by going through the following steps: |
40
|
|
|
|
|
|
|
|
41
|
|
|
|
|
|
|
(0) Choose a dataset to compare everything else to (the middlemost) |
42
|
|
|
|
|
|
|
1) Put the middle of the data at (0,0,0,0...) by subtracting the median ... report the median |
43
|
|
|
|
|
|
|
2) Rotate the data so the line x=y=z=... lies on a single axis. The data should be spread along this axis. |
44
|
|
|
|
|
|
|
3) Do regression on the data and work out "wrongness" of each replicate (!) |
45
|
|
|
|
|
|
|
4) Calculate and report ratio variance and imprecision variance |
46
|
|
|
|
|
|
|
5) Report combined ratio and error for each protein/gene/whatever |
47
|
|
|
|
|
|
|
|
48
|
|
|
|
|
|
|
Perhaps a little code snippet. |
49
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
use Statistics::Reproducibility; |
51
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
my $r = Statistics::Reproducibility->new(); |
53
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
=head1 SUBROUTINES/METHODS |
55
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
=head2 new |
57
|
|
|
|
|
|
|
|
58
|
|
|
|
|
|
|
=cut |
59
|
|
|
|
|
|
|
|
60
|
|
|
|
|
|
|
sub new { |
61
|
0
|
|
|
0
|
1
|
|
my $p = shift; |
62
|
0
|
|
0
|
|
|
|
my $c = ref $p || $p; |
63
|
0
|
|
|
|
|
|
my $o = { |
64
|
|
|
|
|
|
|
# scalars: |
65
|
|
|
|
|
|
|
comparatorIndex => 0, # index of column used to compare |
66
|
|
|
|
|
|
|
k => '', # number of columns |
67
|
|
|
|
|
|
|
n => '', # number of rows |
68
|
|
|
|
|
|
|
vE => '', # variance of "error" (imprecision) |
69
|
|
|
|
|
|
|
vS => '', # variable of experimental spread |
70
|
|
|
|
|
|
|
sdE => '', # s.d. error |
71
|
|
|
|
|
|
|
sdS => '', # s.d. spread |
72
|
|
|
|
|
|
|
derivedFrom => '', # the object derived from |
73
|
|
|
|
|
|
|
derivedReason => '', # the reason the object was derived (e.g. deDiagonalize) |
74
|
|
|
|
|
|
|
|
75
|
|
|
|
|
|
|
# arrays (foreach column) |
76
|
|
|
|
|
|
|
'm' => [], # regression denominator |
77
|
|
|
|
|
|
|
'c' => [], # regression constant |
78
|
|
|
|
|
|
|
# arrays (foreach row) |
79
|
|
|
|
|
|
|
pee => '', # p-value of error |
80
|
|
|
|
|
|
|
pss => '', # p-value of spread |
81
|
|
|
|
|
|
|
pes => '', # p-value of error over spread (??) |
82
|
|
|
|
|
|
|
pse => '', # p-value of spread over error |
83
|
|
|
|
|
|
|
|
84
|
|
|
|
|
|
|
# 2D array (LoL) |
85
|
|
|
|
|
|
|
data => [], |
86
|
|
|
|
|
|
|
|
87
|
|
|
|
|
|
|
#meta info |
88
|
|
|
|
|
|
|
copyOnDerive => [qw/comparatorIndex k n vE vS sdE sdS m c pee pss pes pse copyOnDerive obs/] |
89
|
|
|
|
|
|
|
}; |
90
|
0
|
|
|
|
|
|
bless $o, $c; |
91
|
0
|
|
|
|
|
|
return $o; |
92
|
|
|
|
|
|
|
} |
93
|
|
|
|
|
|
|
|
94
|
|
|
|
|
|
|
=head2 derive |
95
|
|
|
|
|
|
|
|
96
|
|
|
|
|
|
|
derives a new object from an old one... some fields are conserved. |
97
|
|
|
|
|
|
|
Warning: references are copied, so m and c point to the same arrays! |
98
|
|
|
|
|
|
|
However, if you run regression() again, they will point to new arrays. |
99
|
|
|
|
|
|
|
Data is set up with k empty columns. |
100
|
|
|
|
|
|
|
|
101
|
|
|
|
|
|
|
=cut |
102
|
|
|
|
|
|
|
|
103
|
|
|
|
|
|
|
sub derive { |
104
|
0
|
|
|
0
|
1
|
|
my ($o,$reason) = @_; |
105
|
0
|
|
|
|
|
|
my $r = $o->new; |
106
|
0
|
|
|
|
|
|
foreach (@{$o->{copyOnDerive}}){ |
|
0
|
|
|
|
|
|
|
107
|
0
|
|
|
|
|
|
$r->{$_} = $o->{$_}; |
108
|
|
|
|
|
|
|
} |
109
|
0
|
|
|
|
|
|
$r->{derivedFrom} = $o; |
110
|
0
|
|
|
|
|
|
$r->{derivedReason} = $reason; |
111
|
0
|
|
|
|
|
|
$r->{data} = [map {[]} (0..$o->{k}-1)]; |
|
0
|
|
|
|
|
|
|
112
|
0
|
|
|
|
|
|
return $r; |
113
|
|
|
|
|
|
|
} |
114
|
|
|
|
|
|
|
|
115
|
|
|
|
|
|
|
=head2 data |
116
|
|
|
|
|
|
|
|
117
|
|
|
|
|
|
|
Set the data. Should be rectangular, i.e. all columns the same length, and |
118
|
|
|
|
|
|
|
we'll check it is and croak if not... |
119
|
|
|
|
|
|
|
but can contain "empty" cells (empty string), which represent missing values |
120
|
|
|
|
|
|
|
in the data. |
121
|
|
|
|
|
|
|
|
122
|
|
|
|
|
|
|
returns the object for chaining. |
123
|
|
|
|
|
|
|
|
124
|
|
|
|
|
|
|
=cut |
125
|
|
|
|
|
|
|
|
126
|
|
|
|
|
|
|
sub data { |
127
|
0
|
|
|
0
|
1
|
|
my ($o,@columns) = @_; |
128
|
0
|
|
|
|
|
|
$o->{data} = [@columns]; |
129
|
0
|
|
|
|
|
|
$o->{k} = @columns; |
130
|
0
|
|
|
|
|
|
$o->{n} = @{$columns[0]}; |
|
0
|
|
|
|
|
|
|
131
|
0
|
|
|
|
|
|
foreach (1 .. $o->{k}-1){ |
132
|
0
|
|
|
|
|
|
croak "columns different lengths!" |
133
|
0
|
0
|
|
|
|
|
unless @{$columns[$_]} == $o->{n}; |
134
|
|
|
|
|
|
|
} |
135
|
0
|
|
|
|
|
|
return $o; |
136
|
|
|
|
|
|
|
} |
137
|
|
|
|
|
|
|
|
138
|
|
|
|
|
|
|
=head2 run |
139
|
|
|
|
|
|
|
|
140
|
|
|
|
|
|
|
runs a recommended workflow. it's a shortcut for: |
141
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
|
143
|
|
|
|
|
|
|
my $m = $r->subtractMedian(); |
144
|
|
|
|
|
|
|
$m->middlemostColumn(); |
145
|
|
|
|
|
|
|
my $d = $m->deDiagonalize(); |
146
|
|
|
|
|
|
|
$d -> regression(); |
147
|
|
|
|
|
|
|
my $e = $d->rotateToRegressionLine(); |
148
|
|
|
|
|
|
|
$e->variances(); |
149
|
|
|
|
|
|
|
|
150
|
|
|
|
|
|
|
It returns the last object. So you could do: |
151
|
|
|
|
|
|
|
|
152
|
|
|
|
|
|
|
my $results = Statistics::Reproducibility |
153
|
|
|
|
|
|
|
->new() |
154
|
|
|
|
|
|
|
->data($mydata) |
155
|
|
|
|
|
|
|
->run() |
156
|
|
|
|
|
|
|
->printableTable($depth); |
157
|
|
|
|
|
|
|
|
158
|
|
|
|
|
|
|
=cut |
159
|
|
|
|
|
|
|
|
160
|
|
|
|
|
|
|
sub run { |
161
|
0
|
|
|
0
|
1
|
|
my $r = shift; |
162
|
|
|
|
|
|
|
# $r->data([qw/1 2 3 4 5 6 7 8/],[qw/0 1 2 3 4 5 6 7/],[qw/2.1 3.2 4.3 5.4 6.5 7.6 8.7 9.8/]); |
163
|
0
|
|
|
|
|
|
$r->countObservations(); |
164
|
0
|
|
|
|
|
|
$r->regression(); |
165
|
0
|
|
|
|
|
|
my $m = $r->subtractMedian(); |
166
|
0
|
|
|
|
|
|
$m->applyMinimumObservations(2); |
167
|
0
|
|
|
|
|
|
$m->middlemostColumn(); |
168
|
0
|
|
|
|
|
|
my $d = $m->deDiagonalize(); |
169
|
0
|
|
|
|
|
|
$d->applyMinimumObservations(2); |
170
|
0
|
|
|
|
|
|
$d->regression(); |
171
|
0
|
|
|
|
|
|
my $e = $d->rotateToRegressionLine(); |
172
|
0
|
|
|
|
|
|
$e->applyMinimumObservations(2); |
173
|
0
|
|
|
|
|
|
$e->variances(); |
174
|
0
|
|
|
|
|
|
return $e; |
175
|
|
|
|
|
|
|
} |
176
|
|
|
|
|
|
|
|
177
|
|
|
|
|
|
|
=head2 subtractMedian |
178
|
|
|
|
|
|
|
|
179
|
|
|
|
|
|
|
calculates the median for each column, substracts from each column and |
180
|
|
|
|
|
|
|
returns the new object. |
181
|
|
|
|
|
|
|
|
182
|
|
|
|
|
|
|
=cut |
183
|
|
|
|
|
|
|
|
184
|
|
|
|
|
|
|
sub subtractMedian { |
185
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
186
|
0
|
|
|
|
|
|
my $r = $o->derive('subtractMedian'); |
187
|
0
|
0
|
|
|
|
|
my @medians = map {qmedian([map {$_ eq '' ? () : $_} @$_])} @{$o->{data}}; |
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
188
|
0
|
|
|
|
|
|
foreach my $i(0..$#medians){ |
189
|
0
|
0
|
|
|
|
|
$r->{data}->[$i] = [map {$_ eq '' ? '' : $_ - $medians[$i]} @{$o->{data}->[$i]}]; |
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
190
|
|
|
|
|
|
|
} |
191
|
0
|
|
|
|
|
|
$r->{medians} = \@medians; |
192
|
0
|
|
|
|
|
|
return $r; |
193
|
|
|
|
|
|
|
} |
194
|
|
|
|
|
|
|
|
195
|
|
|
|
|
|
|
=head2 middlemostColumn |
196
|
|
|
|
|
|
|
|
197
|
|
|
|
|
|
|
Calculates which of the columns is middlemost and remembers it so all |
198
|
|
|
|
|
|
|
others are compared to it. This can be done instead of using a constructed |
199
|
|
|
|
|
|
|
median dataset as the comparator so that the constructed one does not add to |
200
|
|
|
|
|
|
|
the spread, and does not contribute to the observation count. |
201
|
|
|
|
|
|
|
|
202
|
|
|
|
|
|
|
Note: the method of scoring the columns involves counting which has |
203
|
|
|
|
|
|
|
the most middlemost values. For two columns only, the result will always |
204
|
|
|
|
|
|
|
be the one with the least missing values. I don't think there's anything |
205
|
|
|
|
|
|
|
wrong with that, but just so you know! |
206
|
|
|
|
|
|
|
|
207
|
|
|
|
|
|
|
=cut |
208
|
|
|
|
|
|
|
|
209
|
|
|
|
|
|
|
sub middlemostColumn { |
210
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
211
|
|
|
|
|
|
|
# which is the middle most column? i.e. who has the most medians? |
212
|
|
|
|
|
|
|
|
213
|
0
|
|
|
|
|
|
my @medianCounts = map {0} (1..$o->{k}); # stash counts |
|
0
|
|
|
|
|
|
|
214
|
|
|
|
|
|
|
|
215
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{n}-1){ # each row |
216
|
0
|
|
|
|
|
|
my @row = (); |
217
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{k}-1){ |
218
|
0
|
0
|
0
|
|
|
|
if(defined $o->{data}->[$j]->[$i] |
219
|
|
|
|
|
|
|
&& $o->{data}->[$j]->[$i] ne ''){ |
220
|
0
|
|
|
|
|
|
push @row, $o->{data}->[$j]->[$i]; |
221
|
|
|
|
|
|
|
} |
222
|
|
|
|
|
|
|
} |
223
|
|
|
|
|
|
|
# who's in the middle? |
224
|
0
|
|
|
|
|
|
foreach(medianI(@row)){ |
225
|
0
|
|
|
|
|
|
$medianCounts[$_] ++; |
226
|
|
|
|
|
|
|
} |
227
|
|
|
|
|
|
|
} |
228
|
|
|
|
|
|
|
|
229
|
|
|
|
|
|
|
# which has the most middlemost values? |
230
|
0
|
|
|
|
|
|
my $imax = 0; |
231
|
0
|
|
|
|
|
|
my $max = 0; |
232
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{k}-1){ |
233
|
0
|
0
|
|
|
|
|
if($medianCounts[$i] > $max){ |
234
|
0
|
|
|
|
|
|
$imax = $i; |
235
|
0
|
|
|
|
|
|
$max = $medianCounts[$i]; |
236
|
|
|
|
|
|
|
} |
237
|
|
|
|
|
|
|
} |
238
|
|
|
|
|
|
|
|
239
|
|
|
|
|
|
|
# so now we want to put this column on the left? Or should we just |
240
|
|
|
|
|
|
|
# store that we're going to use this one as the comparator? |
241
|
0
|
|
|
|
|
|
$o->{comparatorIndex} = $imax; |
242
|
0
|
|
|
|
|
|
return $imax; |
243
|
|
|
|
|
|
|
} |
244
|
|
|
|
|
|
|
|
245
|
|
|
|
|
|
|
=head2 constructMedianLeft |
246
|
|
|
|
|
|
|
|
247
|
|
|
|
|
|
|
Make a median column and pop it on the left. Note that the |
248
|
|
|
|
|
|
|
regular median is used here, not the Quick Median estimator. This means |
249
|
|
|
|
|
|
|
that for an even number of observations, the mean of the two middlemost is |
250
|
|
|
|
|
|
|
used, which is not the case for Quick Median. |
251
|
|
|
|
|
|
|
|
252
|
|
|
|
|
|
|
=cut |
253
|
|
|
|
|
|
|
|
254
|
|
|
|
|
|
|
sub constructMedianLeft { |
255
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
256
|
0
|
|
|
|
|
|
my @newcol = (); |
257
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{n}-1){ |
258
|
0
|
|
|
|
|
|
my @row = (); |
259
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{k}-1){ |
260
|
0
|
0
|
0
|
|
|
|
if(defined $o->{data}->[$i]->[$j] |
261
|
|
|
|
|
|
|
&& $o->{data}->[$i]->[$j] ne ''){ |
262
|
0
|
|
|
|
|
|
push @row, $o->{data}->[$i]->[$j]; |
263
|
|
|
|
|
|
|
} |
264
|
|
|
|
|
|
|
} |
265
|
0
|
|
|
|
|
|
push @newcol, median(@row); |
266
|
|
|
|
|
|
|
} |
267
|
0
|
|
|
|
|
|
unshift @{$o->{data}}, \@newcol; |
|
0
|
|
|
|
|
|
|
268
|
0
|
|
|
|
|
|
return \@newcol; |
269
|
|
|
|
|
|
|
} |
270
|
|
|
|
|
|
|
|
271
|
|
|
|
|
|
|
=head2 deDiagonalize |
272
|
|
|
|
|
|
|
|
273
|
|
|
|
|
|
|
Replicated data with some spread will naturally lie along the diagonal line, |
274
|
|
|
|
|
|
|
y=x (in 2 dimensions, or z=y=x... in more). This function aligns the data |
275
|
|
|
|
|
|
|
along one axis by rotation. This is done so that (a) errors are measured |
276
|
|
|
|
|
|
|
approximately perpendicular to the spread of data and (b) unspread data |
277
|
|
|
|
|
|
|
(a ball of points) gives a gradient of zero in Theil Sen estimator, which is |
278
|
|
|
|
|
|
|
correct because if there's no experimental spread then there can be no |
279
|
|
|
|
|
|
|
evidence that the replicates disagree. |
280
|
|
|
|
|
|
|
|
281
|
|
|
|
|
|
|
Note: at this point, any missing values are REPLACED BY ZEROS! This means |
282
|
|
|
|
|
|
|
that these data point will not disagree with any "unchanging" data, but they |
283
|
|
|
|
|
|
|
will not support the reproducibility of "changed" data (data for proteins/genes) |
284
|
|
|
|
|
|
|
that are regulated). The effect of this is that those points will not appear as |
285
|
|
|
|
|
|
|
extreme in the output and will also have a larger error associated with them. |
286
|
|
|
|
|
|
|
|
287
|
|
|
|
|
|
|
A NEW object is returned! comparatorIndex is honoured and conserved, |
288
|
|
|
|
|
|
|
meaning that if you ran middlemostColumn, the result is the column used |
289
|
|
|
|
|
|
|
as the Y axis in all comparisons, and the column itself will contain the |
290
|
|
|
|
|
|
|
experimental variance. |
291
|
|
|
|
|
|
|
|
292
|
|
|
|
|
|
|
=cut |
293
|
|
|
|
|
|
|
|
294
|
|
|
|
|
|
|
sub deDiagonalize { |
295
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
296
|
0
|
|
|
|
|
|
my $r = $o->derive('deDiagonalize'); |
297
|
|
|
|
|
|
|
|
298
|
0
|
|
|
|
|
|
my $ic = $o->{comparatorIndex}; |
299
|
|
|
|
|
|
|
|
300
|
0
|
|
|
|
|
|
my $a = atan2(1,1); |
301
|
|
|
|
|
|
|
|
302
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{k}-1){ |
303
|
0
|
0
|
|
|
|
|
next if $i == $ic; |
304
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{n}-1){ |
305
|
0
|
|
0
|
|
|
|
my $y = $o->{data}->[$i]->[$j] || 0; |
306
|
0
|
|
0
|
|
|
|
my $x = $o->{data}->[$ic]->[$j] || 0; |
307
|
|
|
|
|
|
|
|
308
|
0
|
0
|
0
|
|
|
|
if($y || $x){ |
309
|
0
|
|
|
|
|
|
my $t = atan2($y,$x) - $a; |
310
|
0
|
|
|
|
|
|
my $r = sqrt($x**2 + $y**2); |
311
|
0
|
|
|
|
|
|
($x,$y) = ($r*cos($t), $r*sin($t)); |
312
|
|
|
|
|
|
|
} |
313
|
|
|
|
|
|
|
|
314
|
0
|
|
|
|
|
|
$r->{data}->[$i]->[$j] = $y; |
315
|
0
|
|
|
|
|
|
$r->{data}->[$ic]->[$j] = $x; |
316
|
|
|
|
|
|
|
} |
317
|
|
|
|
|
|
|
|
318
|
|
|
|
|
|
|
# $r->{data}->[$i] = diagonalComponentsN( |
319
|
|
|
|
|
|
|
# $o->{data}->[$i], $o->{data}->[$ic] |
320
|
|
|
|
|
|
|
# ) |
321
|
|
|
|
|
|
|
} |
322
|
|
|
|
|
|
|
|
323
|
|
|
|
|
|
|
#$r->{data}->[$ic] = diagonalDistancesFromOriginN( |
324
|
|
|
|
|
|
|
# $o->{k}, $o->{n}, @{$o->{data}} |
325
|
|
|
|
|
|
|
#); |
326
|
0
|
|
|
|
|
|
return $r; |
327
|
|
|
|
|
|
|
} |
328
|
|
|
|
|
|
|
|
329
|
|
|
|
|
|
|
=head2 countObservations |
330
|
|
|
|
|
|
|
|
331
|
|
|
|
|
|
|
Counts the number of observations present in each point and stores in obs. |
332
|
|
|
|
|
|
|
The result is used by applyMinimumObservations to check for unwanted data |
333
|
|
|
|
|
|
|
before a processing event which turns empties into zeros (like deDiagonalize). |
334
|
|
|
|
|
|
|
|
335
|
|
|
|
|
|
|
=cut |
336
|
|
|
|
|
|
|
|
337
|
|
|
|
|
|
|
sub countObservations { |
338
|
0
|
|
|
0
|
1
|
|
my ($o) = @_; |
339
|
0
|
|
|
|
|
|
my @obs = (); |
340
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{n}-1){ |
341
|
0
|
|
|
|
|
|
my $c = 0; |
342
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{k}-1){ |
343
|
0
|
0
|
0
|
|
|
|
$c++ if defined $o->{data}->[$i]->[$j] && $o->{data}->[$i]->[$j] ne ''; |
344
|
|
|
|
|
|
|
} |
345
|
0
|
|
|
|
|
|
push @obs, $c; |
346
|
|
|
|
|
|
|
} |
347
|
0
|
|
|
|
|
|
$o->{obs} = \@obs; |
348
|
|
|
|
|
|
|
} |
349
|
|
|
|
|
|
|
|
350
|
|
|
|
|
|
|
=head2 applyMinimumObservations |
351
|
|
|
|
|
|
|
|
352
|
|
|
|
|
|
|
A method that blanks any data that does not have a minimum number of |
353
|
|
|
|
|
|
|
values, e.g. if the minimum were 2, the point [2,3,undef] would be fine |
354
|
|
|
|
|
|
|
but [2,undef,undef] would become [undef,undef,undef] |
355
|
|
|
|
|
|
|
|
356
|
|
|
|
|
|
|
=cut |
357
|
|
|
|
|
|
|
|
358
|
|
|
|
|
|
|
sub applyMinimumObservations { |
359
|
0
|
|
|
0
|
1
|
|
my ($o,$min) = @_; |
360
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{n}-1){ |
361
|
0
|
0
|
|
|
|
|
if($o->{obs}->[$j] < $min){ |
362
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{k}-1){ |
363
|
0
|
|
|
|
|
|
$o->{data}->[$i]->[$j] = ''; |
364
|
|
|
|
|
|
|
} |
365
|
0
|
0
|
|
|
|
|
$o->{d}->[$j] = '' if exists $o->{d}; |
366
|
|
|
|
|
|
|
} |
367
|
|
|
|
|
|
|
} |
368
|
|
|
|
|
|
|
} |
369
|
|
|
|
|
|
|
|
370
|
|
|
|
|
|
|
=head2 regression |
371
|
|
|
|
|
|
|
|
372
|
|
|
|
|
|
|
Perform Theil Sen Estimator regression on the data. The regression is |
373
|
|
|
|
|
|
|
done with the comparator on the x axis, but the symmetric parameters |
374
|
|
|
|
|
|
|
are returned for the comparator on the y-axis. |
375
|
|
|
|
|
|
|
|
376
|
|
|
|
|
|
|
=cut |
377
|
|
|
|
|
|
|
|
378
|
|
|
|
|
|
|
sub regression { |
379
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
380
|
0
|
|
|
|
|
|
my @m = map {1} (0..$o->{k}-1); |
|
0
|
|
|
|
|
|
|
381
|
0
|
|
|
|
|
|
my @c = map {0} (0..$o->{k}-1); |
|
0
|
|
|
|
|
|
|
382
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{k}-1){ |
383
|
0
|
0
|
|
|
|
|
next if $i == $o->{comparatorIndex}; |
384
|
0
|
|
|
|
|
|
my ($m,$c) = theilsen( |
385
|
|
|
|
|
|
|
$o->{data}->[$i], |
386
|
|
|
|
|
|
|
$o->{data}->[$o->{comparatorIndex}] |
387
|
|
|
|
|
|
|
); |
388
|
0
|
|
|
|
|
|
$m[$i] = $m; |
389
|
0
|
|
|
|
|
|
$c[$i] = -$c; # - because we're converting to the inverse symmetric |
390
|
|
|
|
|
|
|
} |
391
|
0
|
|
|
|
|
|
$o->{m} = \@m; |
392
|
0
|
|
|
|
|
|
$o->{c} = \@c; |
393
|
0
|
|
|
|
|
|
return ($o->{m}, $o->{c}); |
394
|
|
|
|
|
|
|
} |
395
|
|
|
|
|
|
|
|
396
|
|
|
|
|
|
|
=head2 rotateToRegressionLine |
397
|
|
|
|
|
|
|
|
398
|
|
|
|
|
|
|
do we need this? |
399
|
|
|
|
|
|
|
|
400
|
|
|
|
|
|
|
=cut |
401
|
|
|
|
|
|
|
|
402
|
|
|
|
|
|
|
sub rotateToRegressionLine { |
403
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
404
|
0
|
0
|
|
|
|
|
croak "looks like regression() has not been called on this object" |
405
|
|
|
|
|
|
|
unless defined $o->{c}; |
406
|
|
|
|
|
|
|
|
407
|
|
|
|
|
|
|
# use distanceToLineN([0,0,0],...) to get middle point of line for distance :-) |
408
|
|
|
|
|
|
|
#my $O = [map {0} (1..$o->{k})]; # the origin |
409
|
|
|
|
|
|
|
#my @MC = ($o->{m},$o->{c}); # we'll be using this a lot, maybe |
410
|
|
|
|
|
|
|
|
411
|
|
|
|
|
|
|
#my ($dO,$X) = distanceToLineN($O,@MC); # $X is the "centre" of the line |
412
|
|
|
|
|
|
|
|
413
|
|
|
|
|
|
|
#### |
414
|
0
|
|
|
|
|
|
my $r = $o->derive('rotateToRegressionLine'); |
415
|
|
|
|
|
|
|
|
416
|
0
|
|
|
|
|
|
my $ic = $o->{comparatorIndex}; |
417
|
|
|
|
|
|
|
|
418
|
0
|
|
|
|
|
|
$r->{d} = []; |
419
|
|
|
|
|
|
|
|
420
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{n}-1){ |
421
|
0
|
|
|
|
|
|
foreach my $i(0..$o->{k}-1){ |
422
|
0
|
0
|
|
|
|
|
next if $i == $ic; |
423
|
|
|
|
|
|
|
|
424
|
0
|
|
|
|
|
|
my $m = $o->{m}->[$i]; |
425
|
0
|
|
|
|
|
|
my $c = $o->{c}->[$i]; |
426
|
0
|
|
0
|
|
|
|
my $y = $o->{data}->[$i]->[$j] || $c; |
427
|
0
|
|
0
|
|
|
|
my $x = $o->{data}->[$ic]->[$j] || 0; |
428
|
0
|
|
|
|
|
|
$y -= $c; |
429
|
0
|
0
|
0
|
|
|
|
if($y || $x){ |
430
|
0
|
|
|
|
|
|
my $a = atan2($m,1); |
431
|
0
|
|
|
|
|
|
my $t = atan2($y,$x) - $a; |
432
|
0
|
|
|
|
|
|
my $r = sqrt($x**2 + $y**2); |
433
|
0
|
|
|
|
|
|
($x,$y) = ($r*cos($t), $r*sin($t)); |
434
|
|
|
|
|
|
|
} |
435
|
|
|
|
|
|
|
|
436
|
0
|
|
|
|
|
|
$r->{data}->[$i]->[$j] = $y; |
437
|
0
|
|
|
|
|
|
$r->{data}->[$ic]->[$j] = $x; |
438
|
|
|
|
|
|
|
|
439
|
|
|
|
|
|
|
#my ($d,$x) = distanceToLineN( |
440
|
|
|
|
|
|
|
# [$o->{data}->[$i]->[$j],$o->{data}->[$ic]->[$j]], |
441
|
|
|
|
|
|
|
# [$o->{m}->[$i], $o->{m}->[$ic]], |
442
|
|
|
|
|
|
|
# [$o->{c}->[$i], $o->{c}->[$ic]] |
443
|
|
|
|
|
|
|
#); |
444
|
|
|
|
|
|
|
#my $icv = $o->{data}->[$ic]->[$j] || 0; |
445
|
|
|
|
|
|
|
#my $iv = $o->{data}->[$i]->[$j] || 0; |
446
|
|
|
|
|
|
|
#if($icv < $iv){ |
447
|
|
|
|
|
|
|
# $d *= -1; |
448
|
|
|
|
|
|
|
#} |
449
|
|
|
|
|
|
|
#$r->{data}->[$i]->[$j] = $d; |
450
|
|
|
|
|
|
|
} |
451
|
|
|
|
|
|
|
|
452
|
0
|
|
|
|
|
|
my $sumOfSquares = 0; |
453
|
0
|
|
|
|
|
|
my $sumOfValues = 0; |
454
|
0
|
0
|
|
|
|
|
$sumOfSquares += $_ foreach map {$_ == $ic ? () : $r->{data}->[$_]->[$j] ** 2} (0..$o->{k}-1); |
|
0
|
|
|
|
|
|
|
455
|
0
|
0
|
|
|
|
|
$sumOfValues += $_ foreach map {$_ == $ic ? () : $r->{data}->[$_]->[$j]} (0..$o->{k}-1); |
|
0
|
|
|
|
|
|
|
456
|
0
|
|
|
|
|
|
my $rootsumsquares = sqrt($sumOfSquares); |
457
|
|
|
|
|
|
|
#my ($d,$x) = distanceToLineN( |
458
|
|
|
|
|
|
|
# [@coords], @MC |
459
|
|
|
|
|
|
|
#); |
460
|
|
|
|
|
|
|
# give the distance a sign too! |
461
|
|
|
|
|
|
|
#my $ss = 0; |
462
|
|
|
|
|
|
|
#foreach my $i(0..$r->{k}-1){ |
463
|
|
|
|
|
|
|
# $ss += $r->{data}->[$i]->[$j] * $r->{m}->[$i]; |
464
|
|
|
|
|
|
|
#} |
465
|
|
|
|
|
|
|
|
466
|
0
|
0
|
|
|
|
|
$rootsumsquares *= -1 if $sumOfValues < 0; |
467
|
|
|
|
|
|
|
|
468
|
0
|
|
|
|
|
|
push @{$r->{d}}, $rootsumsquares; # distance to line |
|
0
|
|
|
|
|
|
|
469
|
|
|
|
|
|
|
|
470
|
|
|
|
|
|
|
#my $ss = 0; # sum of squares |
471
|
|
|
|
|
|
|
#foreach my $i(0..$o->{k}-1){ |
472
|
|
|
|
|
|
|
# my $xi = $X->[$i] || 0; |
473
|
|
|
|
|
|
|
# my $di = $o->{data}->[$i]->[$j] || 0; |
474
|
|
|
|
|
|
|
# $ss += ($xi - $di)**2 |
475
|
|
|
|
|
|
|
#} |
476
|
|
|
|
|
|
|
# |
477
|
|
|
|
|
|
|
#$r->{data}->[$ic]->[$j] = sqrt($ss); # distance to center of line |
478
|
|
|
|
|
|
|
|
479
|
|
|
|
|
|
|
} |
480
|
|
|
|
|
|
|
|
481
|
|
|
|
|
|
|
|
482
|
0
|
|
|
|
|
|
return $r; |
483
|
|
|
|
|
|
|
} |
484
|
|
|
|
|
|
|
|
485
|
|
|
|
|
|
|
=head2 variances |
486
|
|
|
|
|
|
|
|
487
|
|
|
|
|
|
|
Calculate variances... i.e. distances from the origin along the line of |
488
|
|
|
|
|
|
|
regression, and distances from the line of regression. This is just like |
489
|
|
|
|
|
|
|
deDiagonalise, except that only two columns are returned. |
490
|
|
|
|
|
|
|
|
491
|
|
|
|
|
|
|
=cut |
492
|
|
|
|
|
|
|
|
493
|
|
|
|
|
|
|
sub variances { |
494
|
0
|
|
|
0
|
1
|
|
my $o = shift; |
495
|
0
|
|
|
|
|
|
my $S; # experimental spread |
496
|
|
|
|
|
|
|
my $E; # imprecision |
497
|
0
|
|
|
|
|
|
my $df = 0; |
498
|
0
|
|
|
|
|
|
my $ic = $o->{comparatorIndex}; |
499
|
|
|
|
|
|
|
# we can give a value for how likely a point is to be there by imprecision alone |
500
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{n}-1){ |
501
|
0
|
0
|
0
|
|
|
|
if($o->{d}->[$j] ne '' && $o->{data}->[$ic]->[$j] ne ''){ |
502
|
0
|
|
|
|
|
|
$E += $o->{d}->[$j] ** 2; |
503
|
0
|
|
|
|
|
|
$S += $o->{data}->[$ic]->[$j] ** 2; |
504
|
0
|
|
|
|
|
|
$df ++; |
505
|
|
|
|
|
|
|
} |
506
|
|
|
|
|
|
|
} |
507
|
0
|
|
|
|
|
|
$E /= $df; |
508
|
0
|
|
|
|
|
|
$S /= $df; |
509
|
0
|
|
|
|
|
|
my $sdE = sqrt($E); |
510
|
0
|
|
|
|
|
|
my $sdS = sqrt($S); |
511
|
0
|
|
|
|
|
|
$o->{vE} = $E; |
512
|
0
|
|
|
|
|
|
$o->{vS} = $S; |
513
|
0
|
|
|
|
|
|
$o->{sdE} = $sdE; |
514
|
0
|
|
|
|
|
|
$o->{sdS} = $sdS; |
515
|
|
|
|
|
|
|
|
516
|
0
|
|
|
|
|
|
$o->{pee} = []; |
517
|
0
|
|
|
|
|
|
$o->{pss} = []; |
518
|
0
|
|
|
|
|
|
$o->{pes} = []; |
519
|
0
|
|
|
|
|
|
$o->{pse} = []; |
520
|
0
|
|
|
|
|
|
foreach my $j(0..$o->{n}-1){ |
521
|
0
|
|
|
|
|
|
my ($pee,$pes,$pss,$pse) = ('','','',''); |
522
|
0
|
0
|
0
|
|
|
|
if($o->{d}->[$j] ne '' && $o->{data}->[$ic]->[$j] ne ''){ |
523
|
0
|
0
|
|
|
|
|
$pee = $sdE ? |
524
|
|
|
|
|
|
|
Statistics::Distributions::tprob ($df,$o->{d}->[$j] / $sdE) |
525
|
|
|
|
|
|
|
: 1; |
526
|
0
|
0
|
|
|
|
|
$pes = $sdS ? |
527
|
|
|
|
|
|
|
Statistics::Distributions::tprob ($df,$o->{d}->[$j] / $sdS) |
528
|
|
|
|
|
|
|
: 1; |
529
|
0
|
0
|
|
|
|
|
$pss = $sdS ? |
530
|
|
|
|
|
|
|
Statistics::Distributions::tprob ($df,$o->{data}->[$ic]->[$j] / $sdS) |
531
|
|
|
|
|
|
|
: 1; |
532
|
0
|
0
|
|
|
|
|
$pse = $sdE ? |
533
|
|
|
|
|
|
|
Statistics::Distributions::tprob ($df,$o->{data}->[$ic]->[$j] / $sdE) |
534
|
|
|
|
|
|
|
: 1; |
535
|
|
|
|
|
|
|
} |
536
|
0
|
|
|
|
|
|
push @{$o->{pee}}, $pee; |
|
0
|
|
|
|
|
|
|
537
|
0
|
|
|
|
|
|
push @{$o->{pes}}, $pes; |
|
0
|
|
|
|
|
|
|
538
|
0
|
|
|
|
|
|
push @{$o->{pss}}, $pss; |
|
0
|
|
|
|
|
|
|
539
|
0
|
|
|
|
|
|
push @{$o->{pse}}, $pse; |
|
0
|
|
|
|
|
|
|
540
|
|
|
|
|
|
|
} |
541
|
0
|
|
|
|
|
|
return ($S,$E); |
542
|
|
|
|
|
|
|
} |
543
|
|
|
|
|
|
|
|
544
|
|
|
|
|
|
|
=head2 printableTable, printTable |
545
|
|
|
|
|
|
|
|
546
|
|
|
|
|
|
|
printableTable returns all available relevant info in a table |
547
|
|
|
|
|
|
|
printTable prints all available relevant info in a table |
548
|
|
|
|
|
|
|
|
549
|
|
|
|
|
|
|
the firts element returned is a list of columns. the rest are the columns. |
550
|
|
|
|
|
|
|
|
551
|
|
|
|
|
|
|
data stored are: |
552
|
|
|
|
|
|
|
|
553
|
|
|
|
|
|
|
# scalars: |
554
|
|
|
|
|
|
|
comparatorIndex # index of column used to compare |
555
|
|
|
|
|
|
|
k |
556
|
|
|
|
|
|
|
n |
557
|
|
|
|
|
|
|
vE # variance of "error" (imprecision) |
558
|
|
|
|
|
|
|
vS # variable of experimental spread |
559
|
|
|
|
|
|
|
sdE # s.d. error |
560
|
|
|
|
|
|
|
sdS # s.d. spread |
561
|
|
|
|
|
|
|
|
562
|
|
|
|
|
|
|
# arrays (foreach column) |
563
|
|
|
|
|
|
|
m # regression denominator |
564
|
|
|
|
|
|
|
c # regression constant |
565
|
|
|
|
|
|
|
# arrays (foreach row) |
566
|
|
|
|
|
|
|
d # distance from regression line |
567
|
|
|
|
|
|
|
pee # p-value of error |
568
|
|
|
|
|
|
|
pss # p-value of spread |
569
|
|
|
|
|
|
|
pes # p-value of error over spread (??) |
570
|
|
|
|
|
|
|
pse # p-value of spread over error |
571
|
|
|
|
|
|
|
|
572
|
|
|
|
|
|
|
# 2D array (LoL) |
573
|
|
|
|
|
|
|
data |
574
|
|
|
|
|
|
|
|
575
|
|
|
|
|
|
|
note that the distance from the center of the distribution |
576
|
|
|
|
|
|
|
is given by the values in data[comparatorIndex] |
577
|
|
|
|
|
|
|
|
578
|
|
|
|
|
|
|
These methods take a single argumen: depth. Every time an object is |
579
|
|
|
|
|
|
|
derived from another (subtractMedian, deDiagonalize and |
580
|
|
|
|
|
|
|
rotateToRegressionLine all do this) the old object is referenced, and |
581
|
|
|
|
|
|
|
you can include the last $depth objects in the output. Set depth to -1 |
582
|
|
|
|
|
|
|
to include all objects. |
583
|
|
|
|
|
|
|
|
584
|
|
|
|
|
|
|
=cut |
585
|
|
|
|
|
|
|
|
586
|
|
|
|
|
|
|
sub printableTable { |
587
|
0
|
|
|
0
|
1
|
|
my ($o,$deep) = @_; |
588
|
0
|
|
|
|
|
|
my @header = (qw/Statistic Value/, map {"Column $_"} (1..$o->{k})); |
|
0
|
|
|
|
|
|
|
589
|
|
|
|
|
|
|
|
590
|
0
|
|
|
|
|
|
my @statistics = (); |
591
|
0
|
|
|
|
|
|
my @values = (); |
592
|
|
|
|
|
|
|
|
593
|
0
|
|
|
|
|
|
my @statkeys = qw/comparatorIndex k n vE vS sdE sdS/; |
594
|
0
|
|
|
|
|
|
my @statnames = qw/CompareColumn Columns Rows ErrorVariance SpreadVariance ErrorSD SpreadSD/; |
595
|
0
|
|
|
|
|
|
foreach (0..$#statkeys){ |
596
|
0
|
|
|
|
|
|
my $statkey = $statkeys[$_]; |
597
|
0
|
|
|
|
|
|
my $statname = $statnames[$_]; |
598
|
0
|
0
|
0
|
|
|
|
if(defined $o->{$statkey} && $o->{$statkey} ne ''){ |
599
|
0
|
|
|
|
|
|
push @statistics, $statname; |
600
|
0
|
|
|
|
|
|
push @values, $o->{$statkey}; |
601
|
0
|
0
|
|
|
|
|
if($statkey eq 'comparatorIndex'){ |
602
|
0
|
|
|
|
|
|
$values[$#values] ++; # 1-based! |
603
|
|
|
|
|
|
|
} |
604
|
|
|
|
|
|
|
} |
605
|
|
|
|
|
|
|
} |
606
|
|
|
|
|
|
|
|
607
|
0
|
|
|
|
|
|
my @printData = (\@statistics, \@values, @{$o->{data}}); |
|
0
|
|
|
|
|
|
|
608
|
|
|
|
|
|
|
|
609
|
0
|
0
|
0
|
|
|
|
if(ref $o->{m} && @{$o->{m}} ){ |
|
0
|
|
|
|
|
|
|
610
|
0
|
|
|
|
|
|
push @header, 'Regression','M','C'; |
611
|
0
|
|
|
|
|
|
push @printData, [map {"Column $_"} (1..$o->{k})]; |
|
0
|
|
|
|
|
|
|
612
|
0
|
|
|
|
|
|
push @printData, $o->{m}, $o->{c}; |
613
|
|
|
|
|
|
|
} |
614
|
|
|
|
|
|
|
|
615
|
0
|
0
|
|
|
|
|
if(ref $o->{d}){ |
616
|
0
|
|
|
|
|
|
push @header, 'DistanceToRegressionLine'; |
617
|
0
|
|
|
|
|
|
push @printData, $o->{d}; |
618
|
|
|
|
|
|
|
} |
619
|
|
|
|
|
|
|
|
620
|
0
|
0
|
|
|
|
|
if(ref $o->{pee}){ |
621
|
0
|
|
|
|
|
|
push @header, 'ErrorPvalue'; |
622
|
0
|
|
|
|
|
|
push @printData, $o->{pee}; |
623
|
|
|
|
|
|
|
|
624
|
0
|
|
|
|
|
|
push @header, 'SpreadPvalue'; |
625
|
0
|
|
|
|
|
|
push @printData, $o->{pss}; |
626
|
|
|
|
|
|
|
|
627
|
0
|
|
|
|
|
|
push @header, 'ErrorOverSpreadPvalue'; |
628
|
0
|
|
|
|
|
|
push @printData, $o->{pes}; |
629
|
|
|
|
|
|
|
|
630
|
0
|
|
|
|
|
|
push @header, 'SpreadOverErrorPvalue'; |
631
|
0
|
|
|
|
|
|
push @printData, $o->{pse}; |
632
|
|
|
|
|
|
|
} |
633
|
|
|
|
|
|
|
|
634
|
0
|
0
|
0
|
|
|
|
if($deep && ref($o->{derivedFrom})){ |
635
|
0
|
|
|
|
|
|
push @header, 'DerivedFrom'; |
636
|
0
|
|
|
|
|
|
push @printData, [$o->{derivedReason}]; |
637
|
0
|
|
|
|
|
|
my $d = $o->{derivedFrom}->printableTable($deep-1); |
638
|
0
|
|
|
|
|
|
my ($dh,@dcols) = @$d; |
639
|
0
|
|
|
|
|
|
push @header, @$dh; |
640
|
0
|
|
|
|
|
|
push @printData, @dcols; |
641
|
|
|
|
|
|
|
} |
642
|
|
|
|
|
|
|
|
643
|
0
|
|
|
|
|
|
return [\@header, @printData]; |
644
|
|
|
|
|
|
|
} |
645
|
|
|
|
|
|
|
|
646
|
|
|
|
|
|
|
sub printTable { |
647
|
0
|
|
|
0
|
1
|
|
my ($o,$deep) = @_; |
648
|
0
|
|
|
|
|
|
my $pt = $o->printableTable($deep); |
649
|
|
|
|
|
|
|
# lets assume that n > number of statistics! |
650
|
0
|
|
|
|
|
|
my $n = $o->{n}; |
651
|
0
|
|
|
|
|
|
my $w = @$pt; |
652
|
0
|
|
|
|
|
|
print join("\t", @{$pt->[0]})."\n"; |
|
0
|
|
|
|
|
|
|
653
|
0
|
|
|
|
|
|
foreach my $j(0..$n-1){ |
654
|
0
|
|
|
|
|
|
my @row = (); |
655
|
0
|
|
|
|
|
|
foreach my $i(1..$w-1){ |
656
|
0
|
0
|
|
|
|
|
my $val = defined $pt->[$i]->[$j] ? $pt->[$i]->[$j] : ''; |
657
|
0
|
|
|
|
|
|
push @row, $val; |
658
|
|
|
|
|
|
|
} |
659
|
0
|
|
|
|
|
|
print join("\t", @row)."\n"; |
660
|
|
|
|
|
|
|
} |
661
|
|
|
|
|
|
|
} |
662
|
|
|
|
|
|
|
|
663
|
|
|
|
|
|
|
=head1 just some wee helper functions... |
664
|
|
|
|
|
|
|
|
665
|
|
|
|
|
|
|
=head2 median |
666
|
|
|
|
|
|
|
|
667
|
|
|
|
|
|
|
yes this probably exists in other modules, but I didn't want to pull in a whole |
668
|
|
|
|
|
|
|
module for just one funciton. Anyway, this is an inefficient version for small |
669
|
|
|
|
|
|
|
numbers of data. It sorts the list and then uses middle() to find the middle of |
670
|
|
|
|
|
|
|
the sorted list. |
671
|
|
|
|
|
|
|
|
672
|
|
|
|
|
|
|
=cut |
673
|
|
|
|
|
|
|
|
674
|
|
|
|
|
|
|
sub median { |
675
|
0
|
0
|
0
|
0
|
1
|
|
my @r = sort {$a<=>$b} map {defined && /\d/ ? $_ : ()} @_; |
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
676
|
0
|
|
|
|
|
|
return middle(@r); |
677
|
|
|
|
|
|
|
} |
678
|
|
|
|
|
|
|
|
679
|
|
|
|
|
|
|
=head2 medianN |
680
|
|
|
|
|
|
|
|
681
|
|
|
|
|
|
|
Like median, but for an even list is returns the two middlemost values. |
682
|
|
|
|
|
|
|
This version is used in medianI. |
683
|
|
|
|
|
|
|
|
684
|
|
|
|
|
|
|
=cut |
685
|
|
|
|
|
|
|
|
686
|
|
|
|
|
|
|
sub medianN { |
687
|
0
|
0
|
0
|
0
|
1
|
|
my @r = sort {$a<=>$b} map {defined && /\d/ ? $_ : ()} @_; |
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
688
|
0
|
|
|
|
|
|
return middleN(@r); |
689
|
|
|
|
|
|
|
} |
690
|
|
|
|
|
|
|
|
691
|
|
|
|
|
|
|
=head2 medianI |
692
|
|
|
|
|
|
|
|
693
|
|
|
|
|
|
|
This uses medianN to get the middlemost value(s) and then returns a list |
694
|
|
|
|
|
|
|
of column indices indicating which columns had a middlemost value. |
695
|
|
|
|
|
|
|
This is used in the medianLeft method when deciding which |
696
|
|
|
|
|
|
|
column is middlemost. |
697
|
|
|
|
|
|
|
|
698
|
|
|
|
|
|
|
=cut |
699
|
|
|
|
|
|
|
|
700
|
|
|
|
|
|
|
sub medianI { |
701
|
0
|
|
|
0
|
1
|
|
my @N = medianN(@_); |
702
|
0
|
|
|
|
|
|
my @I = (); |
703
|
0
|
|
|
|
|
|
foreach my $i(0..$#_){ |
704
|
0
|
0
|
0
|
|
|
|
if(defined $_[$i] && $_[$i] ne ''){ |
705
|
0
|
|
|
|
|
|
foreach my $n(@N){ |
706
|
0
|
0
|
|
|
|
|
if($n == $_[$i]){ |
707
|
0
|
|
|
|
|
|
push @I, $i; |
708
|
|
|
|
|
|
|
} |
709
|
|
|
|
|
|
|
} |
710
|
|
|
|
|
|
|
} |
711
|
|
|
|
|
|
|
} |
712
|
0
|
|
|
|
|
|
return @I; |
713
|
|
|
|
|
|
|
} |
714
|
|
|
|
|
|
|
|
715
|
|
|
|
|
|
|
=head2 middle |
716
|
|
|
|
|
|
|
|
717
|
|
|
|
|
|
|
middle returns the middlemost item in a list, or the mean average of the two |
718
|
|
|
|
|
|
|
middlemost items. It doesn't sort the list first. |
719
|
|
|
|
|
|
|
|
720
|
|
|
|
|
|
|
=cut |
721
|
|
|
|
|
|
|
|
722
|
|
|
|
|
|
|
sub middle { |
723
|
0
|
0
|
|
0
|
1
|
|
if(@_ % 2){ |
724
|
0
|
|
|
|
|
|
return $_[$#_/2]; |
725
|
|
|
|
|
|
|
} |
726
|
|
|
|
|
|
|
else { |
727
|
0
|
|
|
|
|
|
return $_[($#_+1)/2]/2 + $_[($#_-1)/2]/2; |
728
|
|
|
|
|
|
|
} |
729
|
|
|
|
|
|
|
} |
730
|
|
|
|
|
|
|
|
731
|
|
|
|
|
|
|
=head2 middleN |
732
|
|
|
|
|
|
|
|
733
|
|
|
|
|
|
|
middleN does like middle, but for even lists, it returns the two middlemost |
734
|
|
|
|
|
|
|
items as a list. This is used by medianN. |
735
|
|
|
|
|
|
|
|
736
|
|
|
|
|
|
|
=cut |
737
|
|
|
|
|
|
|
|
738
|
|
|
|
|
|
|
sub middleN { |
739
|
0
|
0
|
|
0
|
1
|
|
if(@_ % 2){ |
740
|
0
|
|
|
|
|
|
return $_[$#_/2]; |
741
|
|
|
|
|
|
|
} |
742
|
|
|
|
|
|
|
else { |
743
|
0
|
|
|
|
|
|
return ($_[($#_+1)/2], $_[($#_-1)/2]); |
744
|
|
|
|
|
|
|
} |
745
|
|
|
|
|
|
|
} |
746
|
|
|
|
|
|
|
|
747
|
|
|
|
|
|
|
|
748
|
|
|
|
|
|
|
=head1 AUTHOR |
749
|
|
|
|
|
|
|
|
750
|
|
|
|
|
|
|
Jimi Wills, C<< >> |
751
|
|
|
|
|
|
|
|
752
|
|
|
|
|
|
|
=head1 BUGS |
753
|
|
|
|
|
|
|
|
754
|
|
|
|
|
|
|
Please report any bugs or feature requests to C, or through |
755
|
|
|
|
|
|
|
the web interface at L. I will be notified, and then you'll |
756
|
|
|
|
|
|
|
automatically be notified of progress on your bug as I make changes. |
757
|
|
|
|
|
|
|
|
758
|
|
|
|
|
|
|
|
759
|
|
|
|
|
|
|
|
760
|
|
|
|
|
|
|
|
761
|
|
|
|
|
|
|
=head1 SUPPORT |
762
|
|
|
|
|
|
|
|
763
|
|
|
|
|
|
|
You can find documentation for this module with the perldoc command. |
764
|
|
|
|
|
|
|
|
765
|
|
|
|
|
|
|
perldoc Statistics::Reproducibility |
766
|
|
|
|
|
|
|
|
767
|
|
|
|
|
|
|
|
768
|
|
|
|
|
|
|
You can also look for information at: |
769
|
|
|
|
|
|
|
|
770
|
|
|
|
|
|
|
=over 4 |
771
|
|
|
|
|
|
|
|
772
|
|
|
|
|
|
|
=item * RT: CPAN's request tracker (report bugs here) |
773
|
|
|
|
|
|
|
|
774
|
|
|
|
|
|
|
L |
775
|
|
|
|
|
|
|
|
776
|
|
|
|
|
|
|
=item * AnnoCPAN: Annotated CPAN documentation |
777
|
|
|
|
|
|
|
|
778
|
|
|
|
|
|
|
L |
779
|
|
|
|
|
|
|
|
780
|
|
|
|
|
|
|
=item * CPAN Ratings |
781
|
|
|
|
|
|
|
|
782
|
|
|
|
|
|
|
L |
783
|
|
|
|
|
|
|
|
784
|
|
|
|
|
|
|
=item * Search CPAN |
785
|
|
|
|
|
|
|
|
786
|
|
|
|
|
|
|
L |
787
|
|
|
|
|
|
|
|
788
|
|
|
|
|
|
|
=back |
789
|
|
|
|
|
|
|
|
790
|
|
|
|
|
|
|
|
791
|
|
|
|
|
|
|
=head1 ACKNOWLEDGEMENTS |
792
|
|
|
|
|
|
|
|
793
|
|
|
|
|
|
|
|
794
|
|
|
|
|
|
|
=head1 LICENSE AND COPYRIGHT |
795
|
|
|
|
|
|
|
|
796
|
|
|
|
|
|
|
Copyright 2013 Jimi Wills. |
797
|
|
|
|
|
|
|
|
798
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify it |
799
|
|
|
|
|
|
|
under the terms of the the Artistic License (2.0). You may obtain a |
800
|
|
|
|
|
|
|
copy of the full license at: |
801
|
|
|
|
|
|
|
|
802
|
|
|
|
|
|
|
L |
803
|
|
|
|
|
|
|
|
804
|
|
|
|
|
|
|
Any use, modification, and distribution of the Standard or Modified |
805
|
|
|
|
|
|
|
Versions is governed by this Artistic License. By using, modifying or |
806
|
|
|
|
|
|
|
distributing the Package, you accept this license. Do not use, modify, |
807
|
|
|
|
|
|
|
or distribute the Package, if you do not accept this license. |
808
|
|
|
|
|
|
|
|
809
|
|
|
|
|
|
|
If your Modified Version has been derived from a Modified Version made |
810
|
|
|
|
|
|
|
by someone other than you, you are nevertheless required to ensure that |
811
|
|
|
|
|
|
|
your Modified Version complies with the requirements of this license. |
812
|
|
|
|
|
|
|
|
813
|
|
|
|
|
|
|
This license does not grant you the right to use any trademark, service |
814
|
|
|
|
|
|
|
mark, tradename, or logo of the Copyright Holder. |
815
|
|
|
|
|
|
|
|
816
|
|
|
|
|
|
|
This license includes the non-exclusive, worldwide, free-of-charge |
817
|
|
|
|
|
|
|
patent license to make, have made, use, offer to sell, sell, import and |
818
|
|
|
|
|
|
|
otherwise transfer the Package with respect to any patent claims |
819
|
|
|
|
|
|
|
licensable by the Copyright Holder that are necessarily infringed by the |
820
|
|
|
|
|
|
|
Package. If you institute patent litigation (including a cross-claim or |
821
|
|
|
|
|
|
|
counterclaim) against any party alleging that the Package constitutes |
822
|
|
|
|
|
|
|
direct or contributory patent infringement, then this Artistic License |
823
|
|
|
|
|
|
|
to you shall terminate on the date that such litigation is filed. |
824
|
|
|
|
|
|
|
|
825
|
|
|
|
|
|
|
Disclaimer of Warranty: THE PACKAGE IS PROVIDED BY THE COPYRIGHT HOLDER |
826
|
|
|
|
|
|
|
AND CONTRIBUTORS "AS IS' AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES. |
827
|
|
|
|
|
|
|
THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR |
828
|
|
|
|
|
|
|
PURPOSE, OR NON-INFRINGEMENT ARE DISCLAIMED TO THE EXTENT PERMITTED BY |
829
|
|
|
|
|
|
|
YOUR LOCAL LAW. UNLESS REQUIRED BY LAW, NO COPYRIGHT HOLDER OR |
830
|
|
|
|
|
|
|
CONTRIBUTOR WILL BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, OR |
831
|
|
|
|
|
|
|
CONSEQUENTIAL DAMAGES ARISING IN ANY WAY OUT OF THE USE OF THE PACKAGE, |
832
|
|
|
|
|
|
|
EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
833
|
|
|
|
|
|
|
|
834
|
|
|
|
|
|
|
|
835
|
|
|
|
|
|
|
=cut |
836
|
|
|
|
|
|
|
|
837
|
|
|
|
|
|
|
1; # End of Statistics::Reproducibility |