|
3
|
1 """
|
|
|
2 Author: Timothy Tickle
|
|
|
3 Description: Utility class for generic math functions.
|
|
|
4 """
|
|
|
5
|
|
|
6 #####################################################################################
|
|
|
7 #Copyright (C) <2012>
|
|
|
8 #
|
|
|
9 #Permission is hereby granted, free of charge, to any person obtaining a copy of
|
|
|
10 #this software and associated documentation files (the "Software"), to deal in the
|
|
|
11 #Software without restriction, including without limitation the rights to use, copy,
|
|
|
12 #modify, merge, publish, distribute, sublicense, and/or sell copies of the Software,
|
|
|
13 #and to permit persons to whom the Software is furnished to do so, subject to
|
|
|
14 #the following conditions:
|
|
|
15 #
|
|
|
16 #The above copyright notice and this permission notice shall be included in all copies
|
|
|
17 #or substantial portions of the Software.
|
|
|
18 #
|
|
|
19 #THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
|
|
|
20 #INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
|
|
|
21 #PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
|
|
|
22 #HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
|
|
23 #OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
|
|
24 #SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
|
|
25 #####################################################################################
|
|
|
26
|
|
|
27 __author__ = "Timothy Tickle"
|
|
|
28 __copyright__ = "Copyright 2012"
|
|
|
29 __credits__ = ["Timothy Tickle"]
|
|
|
30 __license__ = "MIT"
|
|
|
31 __maintainer__ = "Timothy Tickle"
|
|
|
32 __email__ = "ttickle@sph.harvard.edu"
|
|
|
33 __status__ = "Development"
|
|
|
34
|
|
|
35 #Import libaries
|
|
|
36 import itertools
|
|
|
37 import numpy as np
|
|
|
38 import operator
|
|
|
39 import random
|
|
|
40 from ValidateData import ValidateData
|
|
|
41
|
|
|
42 class UtilityMath():
|
|
|
43 """
|
|
|
44 Class to perform misc math methods.
|
|
|
45 """
|
|
|
46
|
|
|
47 ##
|
|
|
48 #Happy path test 2
|
|
|
49 @staticmethod
|
|
|
50 def funcConvertToBHQValue(ldPValues, iNumberOfTests=None):
|
|
|
51 """
|
|
|
52 Convert a list of p-value to a list of q-values.
|
|
|
53
|
|
|
54 :param ldPValues: List of doubles (p-values) to convert.
|
|
|
55 :type List
|
|
|
56 :param iNumberOfTests: Number of (multiple) tests if different than the ldValue length. If not set the length of ldPValues is used.
|
|
|
57 :type Integer
|
|
|
58 :return List: List of Q-values made with a BH modification.
|
|
|
59 """
|
|
|
60
|
|
|
61 #If the number of tests is not specified, use the number of pvalues
|
|
|
62 if(iNumberOfTests == None):
|
|
|
63 iNumberOfTests = len(ldPValues)
|
|
|
64 #Used to hold the pvalues as they are being manipulated
|
|
|
65 lsConvertToQValues = list()
|
|
|
66 #Is used to set the ordr of the pvalues as they are placed in the lsConvertToQValues
|
|
|
67 dOrder = 1
|
|
|
68 for dValue in ldPValues:
|
|
|
69 lsConvertToQValues.append([dValue,dOrder,None])
|
|
|
70 dOrder = dOrder + 1
|
|
|
71
|
|
|
72 #Sort by pvalue
|
|
|
73 lsConvertToQValues.sort(key=lambda x: x[0])
|
|
|
74
|
|
|
75 #Used to keep track of the current test number
|
|
|
76 iTest = 1
|
|
|
77 for dConvValue in lsConvertToQValues:
|
|
|
78 dConvValue[2] = dConvValue[0] * iNumberOfTests / iTest
|
|
|
79 iTest = iTest + 1
|
|
|
80
|
|
|
81 #Sort by original order
|
|
|
82 lsConvertToQValues.sort(key=lambda x: x[1])
|
|
|
83
|
|
|
84 #return just 1 dimension (the qvalue)
|
|
|
85 return [ldValues[2] for ldValues in lsConvertToQValues]
|
|
|
86
|
|
|
87 #Happy path tested 5
|
|
|
88 @staticmethod
|
|
|
89 def funcSampleWithReplacement(aData, iSelect):
|
|
|
90 """
|
|
|
91 Sample from a vector of data (aData) with replacement iSelect many objects.
|
|
|
92
|
|
|
93 :param aData: Data to sample from with replacement.
|
|
|
94 :type List
|
|
|
95 :param iSelect: Amount of data to select from the original data population.
|
|
|
96 :type Integer.
|
|
|
97 :return List: List of sampled data.
|
|
|
98 Returns an empty list on error.
|
|
|
99 """
|
|
|
100
|
|
|
101 if iSelect and aData:
|
|
|
102 iDataSize = len(aData)
|
|
|
103 funcRandom, funcInt = random.random, int
|
|
|
104 lsSampling = operator.itemgetter(*[funcInt(funcRandom() * iDataSize) for selected in itertools.repeat(None, iSelect)])(aData)
|
|
|
105 if isinstance(lsSampling, basestring):
|
|
|
106 lsSampling = [lsSampling]
|
|
|
107 return lsSampling
|
|
|
108 return []
|
|
|
109
|
|
|
110 #Happy Path Tested 2
|
|
|
111 @staticmethod
|
|
|
112 def funcSumRowsOfColumns(npaAbundance, lsSampleNames):
|
|
|
113 """
|
|
|
114 Takes the column names of a npArray and sums the rows into one column.
|
|
|
115
|
|
|
116 :param npaAbundance: Array of data to sum.
|
|
|
117 :type Numpy Array
|
|
|
118 :param lsSampleNames: List of sample names.
|
|
|
119 :type List List of strings.
|
|
|
120 :return List List of data summed at each row.
|
|
|
121 """
|
|
|
122
|
|
|
123 #Compress by data name
|
|
|
124 npPooledSample = npaAbundance[lsSampleNames[0]]
|
|
|
125 for strSampleName in lsSampleNames[1:]:
|
|
|
126 #When combining, combine counts by summing
|
|
|
127 npPooledSample = npPooledSample + npaAbundance[strSampleName]
|
|
|
128 return list(npPooledSample)
|
|
|
129
|
|
|
130 #Testing Status: Light happy path testing 2
|
|
|
131 @staticmethod
|
|
|
132 def funcTransposeDataMatrix(npaMatrix, fRemoveAdornments=False):
|
|
|
133 """
|
|
|
134 Transposes a numpy array.
|
|
|
135
|
|
|
136 :param npaMatrix: Data matrix to transpose.
|
|
|
137 :type Numpy Array
|
|
|
138 :param fRemoveAdornments: Remove the first column before transposing.
|
|
|
139 :type Boolean True indicates removing the column.
|
|
|
140 :return Boolean or Numpy Array: Transposed array or a boolean indicating error.
|
|
|
141 Boolean False is returned on error.
|
|
|
142 """
|
|
|
143
|
|
|
144 #Validate parameters
|
|
|
145 if(not ValidateData.funcIsValidNPArray(npaMatrix)):
|
|
|
146 print "".join(["Utility_Math:transposeDataMatrix::Error, transposeDataMatrix was an invalid structured array. Value =",str(npaMatrix)])
|
|
|
147 return False
|
|
|
148 if(not ValidateData.funcIsValidBoolean(fRemoveAdornments)):
|
|
|
149 print "".join(["Utility_Math:transposeDataMatrix::Error, fRemoveAdornments was an invalid boolean. Value =",str(fRemoveAdornments)])
|
|
|
150 return False
|
|
|
151
|
|
|
152 #Change to samples x taxa as is needed for the compute method below
|
|
|
153 #Also remove the first row which is taxa identification
|
|
|
154 conversionMatrix = [list(row)[fRemoveAdornments:] for row in npaMatrix]
|
|
|
155 return np.array(conversionMatrix).transpose()
|
|
|
156
|
