0
|
1 <tool id="secimtools_log_and_glog_transformation" name="Log and Generalized Log (G-Log) Transformation." version="@WRAPPER_VERSION@">
|
|
2 <description></description>
|
|
3 <macros>
|
|
4 <import>macros.xml</import>
|
|
5 </macros>
|
|
6 <expand macro="requirements" />
|
|
7 <command><![CDATA[
|
|
8 log_and_glog_transformation.py
|
|
9 --input $input
|
|
10 --design $design
|
|
11 --uniqID $uniqID
|
|
12 --transformation $transformation
|
|
13 --log_base $log_base
|
|
14 --lambda_value $lambda_value
|
|
15 --oname $oname
|
|
16 ]]></command>
|
|
17 <inputs>
|
|
18 <param name="input" type="data" format="tabular" label="Wide Dataset" help="Input your tab-separated wide format dataset. If file is not tab separated see TIP below."/>
|
|
19 <param name="design" type="data" format="tabular" label="Design File" help="Input your design file (tab-separated). Note you need a 'sampleID' column. If not tab separated see TIP below."/>
|
|
20 <param name="uniqID" size="30" type="text" value="" label="Unique Feature ID" help="Name of the column in your wide dataset that has unique identifiers.."/>
|
|
21 <param name="transformation" size="30" display="radio" type="select" value="log" label="Select Transformation" help="Select log or G-log.">
|
|
22 <option value="log" selected="true">Logarithm</option>
|
|
23 <option value="glog" selected="true">Generalized Logarithm (G-Log)</option>
|
|
24 </param>
|
|
25 <param name="log_base" size="30" type="select" display="radio" value="log" label="Logarithm Base" help="Select logarithm base.">
|
|
26 <option value="log" selected="true">Logarithm base e (natural)</option>
|
|
27 <option value="log2" selected="true">Logarithm base 2</option>
|
|
28 <option value="log10" selected="true">Logarithm base 10</option>
|
|
29 </param>
|
|
30 <param name="lambda_value" size="30" type="text" value="100" label="Regularization Parameter Lambda" help='Regularization parameter lambda is used only for G-log transformation and is ignored for log transformation. Lambda must be non-negative.'/>
|
|
31 </inputs>
|
|
32 <outputs>
|
|
33 <data format="tabular" name="oname" label="${tool.name} on ${on_string}" />
|
|
34 </outputs>
|
|
35 <tests>
|
|
36 <test>
|
|
37 <param name="input" value="ST000006_data.tsv"/>
|
|
38 <param name="design" value="ST000006_design.tsv"/>
|
|
39 <param name="uniqID" value="Retention_Index" />
|
|
40 <param name="transformation" value="glog" />
|
|
41 <param name="log_base" value="log" />
|
|
42 <param name="lambda_value" value="1000000" />
|
|
43 <output name="oname" file="ST000006_log_and_glog_transformation_glog_lambda_1000000.tsv" />
|
|
44 </test>
|
|
45 <test>
|
|
46 <param name="input" value="ST000006_data.tsv"/>
|
|
47 <param name="design" value="ST000006_design.tsv"/>
|
|
48 <param name="uniqID" value="Retention_Index" />
|
|
49 <param name="transformation" value="log" />
|
|
50 <param name="log_base" value="log" />
|
|
51 <param name="lambda_value" value="0" />
|
|
52 <output name="oname" file="ST000006_log_and_glog_transformation_log.tsv" />
|
|
53 </test>
|
|
54 </tests>
|
|
55 <help><![CDATA[
|
|
56
|
|
57 @TIP_AND_WARNING@
|
|
58
|
|
59 **Tool Description**
|
|
60
|
|
61 ***NOTE: Zero or negative values in the original dataset will be replaced with missing values after log transformation since logarithms are not defined for non-positive values.***
|
|
62
|
|
63 This tool carries out either log or generalized log (G-log) transformation of values in a Wide Format dataset using the base specified by the user.
|
|
64
|
|
65 The logarithmic transformation has the formula: log(data).
|
|
66
|
|
67 The generalized logarithmic transformation has the formula: log(data + sqrt(data^2 + lambda)).
|
|
68
|
|
69 The generalized version becomes the standard logarithmic transformation re-scaled by sqrt(2) if the lambda value is 0.
|
|
70
|
|
71 Three bases are available for both logarithmic transformations:
|
|
72
|
|
73 base e (natural), base 2, and base 10.
|
|
74
|
|
75
|
|
76
|
|
77 --------------------------------------------------------------------------------
|
|
78
|
|
79 **Input**
|
|
80
|
|
81 - Two input datasets are required.
|
|
82
|
|
83 @WIDE@
|
|
84
|
|
85 **NOTE:** The sample IDs must match the sample IDs in the Design File (below).
|
|
86 Extra columns will automatically be ignored.
|
|
87
|
|
88 @METADATA@
|
|
89
|
|
90 @UNIQID@
|
|
91
|
|
92 **Transformation Choice**
|
|
93
|
|
94 - User has to choose between logarithmic and generalized logarithmic (G-log) transformation.
|
|
95
|
|
96 **Logarithm Base**
|
|
97
|
|
98 - Select base of the logarithm.
|
|
99
|
|
100 **Regularization Parameter Lambda**
|
|
101
|
|
102 - Enter a value for regularization parameter lambda. The value must to be non-negative and is used only for G-log transformation. Default = 100.
|
|
103
|
|
104 --------------------------------------------------------------------------------
|
|
105
|
|
106 **Output**
|
|
107
|
|
108 A TSV file containing the same column names as the original Wide Dataset where the values in each cell correspond to the values obtained by the selected log transformation procedure.
|
|
109
|
|
110 ***NOTE:*** If the original dataset contains 0 or negative values, they will be replaced with missing values after log transformation since logarithms are not defined for non-positive values.
|
|
111 Any values missing in the original dataset will remain missing.
|
|
112
|
|
113 ]]></help>
|
|
114 <expand macro="citations"/>
|
|
115 </tool>
|