|
0
|
1 <tool id="consensus_clustering" name="Consensus Clustering" force_history_refresh="True">
|
|
|
2 <command interpreter="python">consensus.clustering.py
|
|
|
3 -d $dataset
|
|
|
4 -n ${direction}
|
|
|
5 -a ${method.algorithm}
|
|
|
6 #if $method.algorithm == 'hc' # -m ${method.hc_distance_metric}
|
|
|
7 -i ${method.innerLinkage}
|
|
|
8 #end if
|
|
|
9 #if $method.algorithm == 'pam' # -m ${method.pam_distance_metric}
|
|
|
10 #end if
|
|
|
11 #if $method.algorithm == 'km' # -m euclidean
|
|
|
12 #end if
|
|
|
13 -k ${kmax}
|
|
|
14 -r ${reps}
|
|
|
15 -f ${finalLinkage}
|
|
8
|
16 #if str($direction) == "rows":
|
|
|
17 -o ${rdata_output_rows}
|
|
|
18 #end if
|
|
|
19
|
|
|
20 #if str($direction) == "cols":
|
|
|
21 -o ${rdata_output_cols}
|
|
|
22 #end if
|
|
0
|
23 -h $report
|
|
|
24 -p ${report.files_path}
|
|
|
25
|
|
|
26 </command>
|
|
|
27 <inputs>
|
|
|
28 <param name="dataset" type="data" format='tabular' label="Data Set" help="Specify the data matrix (tab-delimited) to be clustered"/>
|
|
|
29 <param name="direction" type="select" label="Cluster Samples or Genes?" help="Specify the matrix dimension to cluster (see help below)">
|
|
|
30 <option value="rows">Genes (rows)</option>
|
|
|
31 <option value="cols" selected="true">Samples (columns)</option>
|
|
|
32 </param>
|
|
|
33
|
|
|
34 <conditional name='method'>
|
|
|
35 <param name="algorithm" type="select" label="Clustering Algorithm" help="Specify the cluster method to use (see help below)">
|
|
|
36 <option value="hc">Hierarchical Clustering</option>
|
|
|
37 <option value="pam" selected='true'>Partioning around Medioids</option>
|
|
|
38 <option value="km">K-Means Clustering</option>
|
|
|
39 </param>
|
|
|
40 <when value='hc'>
|
|
|
41 <param name="hc_distance_metric" type="select" label="Distance Metric" help="Specify the distance metric to use (see help below)">
|
|
|
42 <option value="cosine" selected='true'>Cosine</option>
|
|
|
43 <option value="abscosine">Absolute Cosine</option>
|
|
|
44 <option value="pearson">Pearson</option>
|
|
|
45 <option value="abspearson">Absolute Pearson</option>
|
|
|
46 <option value="spearman">Spearman</option>
|
|
|
47 <option value="kendall">Kendall</option>
|
|
|
48 <option value="euclidean">Euclidean</option>
|
|
|
49 <option value="maximum">Maximum</option>
|
|
|
50 <option value="manhattan">Manhattan (AKA city block)</option>
|
|
|
51 <option value="canberra">Canberra</option>
|
|
|
52 <option value="binary">Binary</option>
|
|
|
53 </param>
|
|
|
54
|
|
|
55 <param name="innerLinkage" type="select" label="Linkage for inner HAC " help="Specify the linkage to use during the 'inner' hierarchcial clustering (see help below)">
|
|
|
56 <option value="average">Average</option>
|
|
|
57 <option value="centroid">Centroid</option>
|
|
|
58 <option value="complete" selected='true'>Complete</option>
|
|
|
59 <option value="mcquitty">McQuitty</option>
|
|
|
60 <option value="median">Median</option>
|
|
|
61 <option value="single">Single</option>
|
|
|
62 <option value="ward">Ward</option>
|
|
|
63 </param>
|
|
|
64 </when>
|
|
|
65 <when value='pam'>
|
|
|
66 <param name="pam_distance_metric" type="select" label="Distance Metric" help="Specify the distance metric to use (see help below)">
|
|
|
67 <option value="cosine" selected='true'>Cosine</option>
|
|
|
68 <option value="abscosine">Absolute Cosine</option>
|
|
|
69 <option value="pearson">Pearson</option>
|
|
|
70 <option value="abspearson">Absolute Pearson</option>
|
|
|
71 <option value="spearman">Spearman</option>
|
|
|
72 <option value="kendall">Kendall</option>
|
|
|
73 <option value="euclidean">Euclidean</option>
|
|
|
74 <option value="maximum">Maximum</option>
|
|
|
75 <option value="manhattan">Manhattan (AKA city block)</option>
|
|
|
76 <option value="canberra">Canberra</option>
|
|
|
77 <option value="binary">Binary</option>
|
|
|
78 </param>
|
|
|
79 </when>
|
|
|
80 </conditional>
|
|
|
81 <param name="finalLinkage" type="select" label="Final Linkage" help="Specify the linkage to use when clustering the consensus matrix (see help below)">
|
|
|
82 <option value="average">Average</option>
|
|
|
83 <option value="centroid">Centroid</option>
|
|
|
84 <option value="complete" selected='true'>Complete</option>
|
|
|
85 <option value="mcquitty">McQuitty</option>
|
|
|
86 <option value="median">Median</option>
|
|
|
87 <option value="single">Single</option>
|
|
|
88 <option value="ward">Ward</option>
|
|
|
89 </param>
|
|
|
90
|
|
|
91
|
|
|
92 <param name="kmax" type="integer" label="K Max" value="10" help="Maximum number of K to analyze" />
|
|
|
93 <param name="reps" type="integer" label="Repetitions" value="500" help="Number of Sample Permutations to Repeat"/>
|
|
|
94
|
|
|
95 </inputs>
|
|
|
96 <outputs>
|
|
|
97 <data format="html" name="report" label="Consensus Clustering Report (HTML)"/>
|
|
8
|
98 <data format="rdata" name="rdata_output_rows" label="Consensus Clustering Results; Gene Clusters (RData)">
|
|
|
99 <filter>(direction)=="rows"</filter>
|
|
|
100 </data>
|
|
|
101 <data format="rdata" name="rdata_output_cols" label="Consensus Clustering Results; Sample Clusters (RData)">
|
|
|
102 <filter>(direction)=="cols"</filter>
|
|
|
103 </data>
|
|
0
|
104 </outputs>
|
|
|
105 <help>
|
|
|
106 .. class:: infomark
|
|
|
107
|
|
|
108 **Perform Consensus Clustering (Cluster Samples) on a specified data set**
|
|
|
109
|
|
|
110 ----
|
|
|
111
|
|
|
112 **Parameters**
|
|
|
113
|
|
|
114 - **Data Set** - Specify the data matrix to be clustered. Data must be formated as follows:
|
|
|
115
|
|
|
116 * Tab-delimited
|
|
|
117 * Use row/column headers
|
|
|
118
|
|
|
119 - **Cluster Samples or Genes** - Specify the dimension of the matrix to cluster:
|
|
|
120
|
|
|
121 * Rows (Genes)
|
|
|
122 * Columns (Samples)
|
|
|
123
|
|
|
124 - **Clustering Algorithm** Specify the choice of algorithm to use. Choice of:
|
|
|
125
|
|
|
126 * Hierarchical Clustering
|
|
|
127 * K-Means
|
|
|
128
|
|
|
129 - **Distance Metric** Specify the choice of distance metric to use. Choice of:
|
|
|
130
|
|
|
131 * Cosine (AKA uncentered pearson)
|
|
|
132 * Absolute Cosine (AKA uncentered pearson, absolute value)
|
|
|
133 * Pearson (pearson correlation)
|
|
|
134 * Absolute Pearson (pearson correlation, absolute value)
|
|
|
135 * Spearman (spearman correlation)
|
|
|
136 * Kendall (Kendall's Tau)
|
|
|
137 * Euclidean (euclidean distance)
|
|
|
138 * Maximum
|
|
|
139 * Manhattan (AKA city block)
|
|
|
140 * Canberra
|
|
|
141 * Binary
|
|
|
142
|
|
|
143 - **Final Linkage** Specify the choice linkage to use when clustering Consensus Matrix. Choice of:
|
|
|
144
|
|
|
145 * Average (see documentation for R's hclust function for explanation of choices)
|
|
|
146 * Single
|
|
|
147 * Complete
|
|
|
148 * Median
|
|
|
149 * Centroid
|
|
|
150 * McQuity
|
|
|
151 * Ward
|
|
|
152
|
|
|
153 - **Inner Linkage** Specify the choice linkage to use when using HAC as clustering method. Same choices as 'Final Linkage'
|
|
|
154
|
|
|
155 - **K Max** Specify the number to use for the largest K considered
|
|
|
156
|
|
|
157 - **Repititions** Specify the number of 'bootstrap' repitions to perform to generate the consensus matrix
|
|
|
158
|
|
|
159 </help>
|
|
|
160 </tool>
|
