annotate rgedgeRpaired_nocamera.xml @ 38:2f7573aacac3 draft

Uploaded
author fubar
date Sun, 22 Dec 2013 02:57:05 -0500
parents 8051d14cf819
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
32
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1 <tool id="rgDifferentialCount" name="Differential_Count" version="0.22">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
2 <description>models using BioConductor packages</description>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
3 <requirements>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
4 <requirement type="package" version="3.0.1">r3</requirement>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
5 <requirement type="package" version="1.3.18">graphicsmagick</requirement>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
6 <requirement type="package" version="9.07">ghostscript</requirement>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
7 <requirement type="package" version="2.12">biocbasics</requirement>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
8 </requirements>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
9
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
10 <command interpreter="python">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
11 rgToolFactory.py --script_path "$runme" --interpreter "Rscript" --tool_name "DifferentialCounts"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
12 --output_dir "$html_file.files_path" --output_html "$html_file" --make_HTML "yes"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
13 </command>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
14 <inputs>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
15 <param name="input1" type="data" format="tabular" label="Select an input matrix - rows are contigs, columns are counts for each sample"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
16 help="Use the HTSeq based count matrix preparation tool to create these matrices from BAM/SAM files and a GTF file of genomic features"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
17 <param name="title" type="text" value="Differential Counts" size="80" label="Title for job outputs"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
18 help="Supply a meaningful name here to remind you what the outputs contain">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
19 <sanitizer invalid_char="">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
20 <valid initial="string.letters,string.digits"><add value="_" /> </valid>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
21 </sanitizer>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
22 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
23 <param name="treatment_name" type="text" value="Treatment" size="50" label="Treatment Name"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
24 <param name="Treat_cols" label="Select columns containing treatment." type="data_column" data_ref="input1" numerical="True"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
25 multiple="true" use_header_names="true" size="120" display="checkboxes">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
26 <validator type="no_options" message="Please select at least one column."/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
27 <sanitizer invalid_char="">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
28 <valid initial="string.letters,string.digits"><add value="_" /> </valid>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
29 </sanitizer>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
30 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
31 <param name="control_name" type="text" value="Control" size="50" label="Control Name"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
32 <param name="Control_cols" label="Select columns containing control." type="data_column" data_ref="input1" numerical="True"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
33 multiple="true" use_header_names="true" size="120" display="checkboxes" optional="true">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
34 <validator type="no_options" message="Please select at least one column."/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
35 <sanitizer invalid_char="">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
36 <valid initial="string.letters,string.digits"><add value="_" /> </valid>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
37 </sanitizer> <validator type="no_options" message="Please select at least one column."/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
38 <sanitizer invalid_char="">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
39 <valid initial="string.letters,string.digits"><add value="_" /> </valid>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
40 </sanitizer>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
41
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
42 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
43 <param name="subjectids" type="text" optional="true" size="120" value = ""
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
44 label="IF SUBJECTS NOT ALL INDEPENDENT! Enter comma separated strings to indicate sample labels for (eg) pairing - must be one for every column in input"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
45 help="Leave blank if no pairing, but eg if data from sample id A99 is in columns 2,4 and id C21 is in 3,5 then enter 'A99,C21,A99,C21'">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
46 <sanitizer>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
47 <valid initial="string.letters,string.digits"><add value="," /> </valid>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
48 </sanitizer>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
49 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
50 <param name="fQ" type="float" value="0.3" size="5" label="Non-differential contig count quantile threshold - zero to analyze all non-zero read count contigs"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
51 help="May be a good or a bad idea depending on the biology and the question. EG 0.3 = sparsest 30% of contigs with at least one read are removed before analysis"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
52 <param name="useNDF" type="boolean" truevalue="T" falsevalue="F" checked="false" size="1"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
53 label="Non differential filter - remove contigs below a threshold (1 per million) for half or more samples"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
54 help="May be a good or a bad idea depending on the biology and the question. This was the old default. Quantile based is available as an alternative"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
55
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
56 <conditional name="edgeR">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
57 <param name="doedgeR" type="select"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
58 label="Run this model using edgeR"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
59 help="edgeR uses a negative binomial model and seems to be powerful, even with few replicates">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
60 <option value="F">Do not run edgeR</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
61 <option value="T" selected="true">Run edgeR</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
62 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
63 <when value="T">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
64 <param name="edgeR_priordf" type="integer" value="20" size="3"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
65 label="prior.df for tagwise dispersion - lower value = more emphasis on each tag's variance. Replaces prior.n and prior.df = prior.n * residual.df"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
66 help="0 = Use edgeR default. Use a small value to 'smooth' small samples. See edgeR docs and note below"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
67 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
68 <when value="F"></when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
69 </conditional>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
70 <conditional name="DESeq2">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
71 <param name="doDESeq2" type="select"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
72 label="Run the same model with DESeq2 and compare findings"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
73 help="DESeq2 is an update to the DESeq package. It uses different assumptions and methods to edgeR">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
74 <option value="F" selected="true">Do not run DESeq2</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
75 <option value="T">Run DESeq2</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
76 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
77 <when value="T">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
78 <param name="DESeq_fitType" type="select">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
79 <option value="parametric" selected="true">Parametric (default) fit for dispersions</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
80 <option value="local">Local fit - this will automagically be used if parametric fit fails</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
81 <option value="mean">Mean dispersion fit- use this if you really understand what you're doing - read the fine manual linked below in the documentation</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
82 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
83 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
84 <when value="F"> </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
85 </conditional>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
86 <param name="doVoom" type="select"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
87 label="Run the same model with Voom/limma and compare findings"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
88 help="Voom uses counts per million and a precise transformation of variance so count data can be analysed using limma">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
89 <option value="F" selected="true">Do not run VOOM</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
90 <option value="T">Run VOOM</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
91 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
92 <!--
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
93 <conditional name="camera">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
94 <param name="doCamera" type="select" label="Run the edgeR implementation of Camera GSEA for up/down gene sets"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
95 help="If yes, you can choose a set of genesets to test and/or supply a gmt format geneset collection from your history">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
96 <option value="F" selected="true">Do not run GSEA tests with the Camera algorithm</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
97 <option value="T">Run GSEA tests with the Camera algorithm</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
98 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
99 <when value="T">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
100 <conditional name="gmtSource">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
101 <param name="refgmtSource" type="select"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
102 label="Use a gene set (.gmt) from your history and/or use a built-in (MSigDB etc) gene set">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
103 <option value="indexed" selected="true">Use a built-in gene set</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
104 <option value="history">Use a gene set from my history</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
105 <option value="both">Add a gene set from my history to a built in gene set</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
106 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
107 <when value="indexed">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
108 <param name="builtinGMT" type="select" label="Select a gene set matrix (.gmt) file to use for the analysis">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
109 <options from_data_table="gseaGMT_3.1">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
110 <filter type="sort_by" column="2" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
111 <validator type="no_options" message="No GMT v3.1 files are available - please install them"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
112 </options>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
113 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
114 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
115 <when value="history">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
116 <param name="ownGMT" type="data" format="gmt" label="Select a Gene Set from your history" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
117 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
118 <when value="both">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
119 <param name="ownGMT" type="data" format="gseagmt" label="Select a Gene Set from your history" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
120 <param name="builtinGMT" type="select" label="Select a gene set matrix (.gmt) file to use for the analysis">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
121 <options from_data_table="gseaGMT_4">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
122 <filter type="sort_by" column="2" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
123 <validator type="no_options" message="No GMT v4 files are available - please fix tool_data_table and loc files"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
124 </options>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
125 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
126 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
127 </conditional>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
128 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
129 <when value="F">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
130 </when>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
131 </conditional>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
132 -->
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
133 <param name="fdrthresh" type="float" value="0.05" size="5" label="P value threshold for FDR filtering for amily wise error rate control"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
134 help="Conventional default value of 0.05 recommended"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
135 <param name="fdrtype" type="select" label="FDR (Type II error) control method"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
136 help="Use fdr or bh typically to control for the number of tests in a reliable way">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
137 <option value="fdr" selected="true">fdr</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
138 <option value="BH">Benjamini Hochberg</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
139 <option value="BY">Benjamini Yukateli</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
140 <option value="bonferroni">Bonferroni</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
141 <option value="hochberg">Hochberg</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
142 <option value="holm">Holm</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
143 <option value="hommel">Hommel</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
144 <option value="none">no control for multiple tests</option>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
145 </param>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
146 </inputs>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
147 <outputs>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
148 <data format="tabular" name="out_edgeR" label="${title}_topTable_edgeR.xls">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
149 <filter>edgeR['doedgeR'] == "T"</filter>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
150 </data>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
151 <data format="tabular" name="out_DESeq2" label="${title}_topTable_DESeq2.xls">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
152 <filter>DESeq2['doDESeq2'] == "T"</filter>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
153 </data>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
154 <data format="tabular" name="out_VOOM" label="${title}_topTable_VOOM.xls">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
155 <filter>doVoom == "T"</filter>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
156 </data>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
157 <data format="html" name="html_file" label="${title}.html"/>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
158 </outputs>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
159 <stdio>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
160 <exit_code range="4" level="fatal" description="Number of subject ids must match total number of samples in the input matrix" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
161 </stdio>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
162 <tests>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
163 <test>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
164 <param name='input1' value='test_bams2mx.xls' ftype='tabular' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
165 <param name='treatment_name' value='liver' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
166 <param name='title' value='edgeRtest' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
167 <param name='useNDF' value='' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
168 <param name='doedgeR' value='T' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
169 <param name='doVoom' value='T' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
170 <param name='doDESeq2' value='T' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
171 <param name='fdrtype' value='fdr' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
172 <param name='edgeR_priordf' value="8" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
173 <param name='fdrthresh' value="0.05" />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
174 <param name='control_name' value='heart' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
175 <param name='subjectids' value='' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
176 <param name='Control_cols' value='3,4,5,9' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
177 <param name='Treat_cols' value='2,6,7,8' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
178 <output name='out_edgeR' file='edgeRtest1out.xls' compare='diff' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
179 <output name='html_file' file='edgeRtest1out.html' compare='diff' lines_diff='20' />
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
180 </test>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
181 </tests>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
182
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
183 <configfiles>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
184 <configfile name="runme">
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
185 <![CDATA[
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
186 #
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
187 # edgeR.Rscript
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
188 # updated npv 2011 for R 2.14.0 and edgeR 2.4.0 by ross
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
189 # Performs DGE on a count table containing n replicates of two conditions
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
190 #
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
191 # Parameters
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
192 #
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
193 # 1 - Output Dir
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
194
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
195 # Original edgeR code by: S.Lunke and A.Kaspi
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
196 reallybig = log10(.Machine\$double.xmax)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
197 reallysmall = log10(.Machine\$double.xmin)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
198 library('stringr')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
199 library('gplots')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
200 library('edgeR')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
201 hmap2 = function(cmat,nsamp=100,outpdfname='heatmap2.pdf', TName='Treatment',group=NA,myTitle='title goes here')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
202 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
203 # Perform clustering for significant pvalues after controlling FWER
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
204 samples = colnames(cmat)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
205 gu = unique(group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
206 gn = rownames(cmat)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
207 if (length(gu) == 2) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
208 col.map = function(g) {if (g==gu[1]) "#FF0000" else "#0000FF"}
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
209 pcols = unlist(lapply(group,col.map))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
210 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
211 colours = rainbow(length(gu),start=0,end=4/6)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
212 pcols = colours[match(group,gu)] }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
213 dm = cmat[(! is.na(gn)),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
214 # remove unlabelled hm rows
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
215 nprobes = nrow(dm)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
216 # sub = paste('Showing',nprobes,'contigs ranked for evidence of differential abundance')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
217 if (nprobes > nsamp) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
218 dm =dm[1:nsamp,]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
219 #sub = paste('Showing',nsamp,'contigs ranked for evidence for differential abundance out of',nprobes,'total')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
220 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
221 newcolnames = substr(colnames(dm),1,20)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
222 colnames(dm) = newcolnames
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
223 pdf(outpdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
224 heatmap.2(dm,main=myTitle,ColSideColors=pcols,col=topo.colors(100),dendrogram="col",key=T,density.info='none',
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
225 Rowv=F,scale='row',trace='none',margins=c(8,8),cexRow=0.4,cexCol=0.5)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
226 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
227 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
228
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
229 hmap = function(cmat,nmeans=4,outpdfname="heatMap.pdf",nsamp=250,TName='Treatment',group=NA,myTitle="Title goes here")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
230 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
231 # for 2 groups only was
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
232 #col.map = function(g) {if (g==TName) "#FF0000" else "#0000FF"}
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
233 #pcols = unlist(lapply(group,col.map))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
234 gu = unique(group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
235 colours = rainbow(length(gu),start=0.3,end=0.6)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
236 pcols = colours[match(group,gu)]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
237 nrows = nrow(cmat)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
238 mtitle = paste(myTitle,'Heatmap: n contigs =',nrows)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
239 if (nrows > nsamp) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
240 cmat = cmat[c(1:nsamp),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
241 mtitle = paste('Heatmap: Top ',nsamp,' DE contigs (of ',nrows,')',sep='')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
242 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
243 newcolnames = substr(colnames(cmat),1,20)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
244 colnames(cmat) = newcolnames
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
245 pdf(outpdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
246 heatmap(cmat,scale='row',main=mtitle,cexRow=0.3,cexCol=0.4,Rowv=NA,ColSideColors=pcols)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
247 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
248 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
249
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
250 qqPlot = function(descr='qqplot',pvector, outpdf='qqplot.pdf',...)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
251 # stolen from https://gist.github.com/703512
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
252 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
253 o = -log10(sort(pvector,decreasing=F))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
254 e = -log10( 1:length(o)/length(o) )
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
255 o[o==-Inf] = reallysmall
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
256 o[o==Inf] = reallybig
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
257 maint = descr
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
258 pdf(outpdf)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
259 plot(e,o,pch=19,cex=1, main=maint, ...,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
260 xlab=expression(Expected~~-log[10](italic(p))),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
261 ylab=expression(Observed~~-log[10](italic(p))),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
262 xlim=c(0,max(e)), ylim=c(0,max(o)))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
263 lines(e,e,col="red")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
264 grid(col = "lightgray", lty = "dotted")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
265 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
266 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
267
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
268 smearPlot = function(DGEList,deTags, outSmear, outMain)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
269 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
270 pdf(outSmear)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
271 plotSmear(DGEList,de.tags=deTags,main=outMain)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
272 grid(col="lightgray", lty="dotted")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
273 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
274 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
275
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
276 boxPlot = function(rawrs,cleanrs,maint,myTitle,pdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
277 { #
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
278 nc = ncol(rawrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
279 for (i in c(1:nc)) {rawrs[(rawrs[,i] < 0),i] = NA}
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
280 fullnames = colnames(rawrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
281 newcolnames = substr(colnames(rawrs),1,20)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
282 colnames(rawrs) = newcolnames
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
283 newcolnames = substr(colnames(cleanrs),1,20)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
284 colnames(cleanrs) = newcolnames
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
285 defpar = par(no.readonly=T)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
286 print.noquote('raw contig counts by sample:')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
287 print.noquote(summary(rawrs))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
288 print.noquote('normalised contig counts by sample:')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
289 print.noquote(summary(cleanrs))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
290 pdf(pdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
291 par(mfrow=c(1,2))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
292 boxplot(rawrs,varwidth=T,notch=T,ylab='log contig count',col="maroon",las=3,cex.axis=0.35,main=paste('Raw:',maint))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
293 grid(col="lightgray",lty="dotted")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
294 boxplot(cleanrs,varwidth=T,notch=T,ylab='log contig count',col="maroon",las=3,cex.axis=0.35,main=paste('After ',maint))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
295 grid(col="lightgray",lty="dotted")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
296 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
297 pdfname = "sample_counts_histogram.pdf"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
298 nc = ncol(rawrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
299 print.noquote(paste('Using ncol rawrs=',nc))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
300 ncroot = round(sqrt(nc))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
301 if (ncroot*ncroot < nc) { ncroot = ncroot + 1 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
302 m = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
303 for (i in c(1:nc)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
304 rhist = hist(rawrs[,i],breaks=100,plot=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
305 m = append(m,max(rhist\$counts))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
306 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
307 ymax = max(m)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
308 ncols = length(fullnames)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
309 if (ncols > 20)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
310 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
311 scale = 7*ncols/20
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
312 pdf(pdfname,width=scale,height=scale)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
313 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
314 pdf(pdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
315 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
316 par(mfrow=c(ncroot,ncroot))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
317 for (i in c(1:nc)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
318 hist(rawrs[,i], main=paste("Contig logcount",i), xlab='log raw count', col="maroon",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
319 breaks=100,sub=fullnames[i],cex=0.8,ylim=c(0,ymax))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
320 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
321 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
322 par(defpar)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
323
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
324 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
325
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
326 cumPlot = function(rawrs,cleanrs,maint,myTitle)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
327 { # updated to use ecdf
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
328 pdfname = "Filtering_rowsum_bar_charts.pdf"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
329 defpar = par(no.readonly=T)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
330 lrs = log(rawrs,10)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
331 lim = max(lrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
332 pdf(pdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
333 par(mfrow=c(2,1))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
334 hist(lrs,breaks=100,main=paste('Before:',maint),xlab="# Reads (log)",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
335 ylab="Count",col="maroon",sub=myTitle, xlim=c(0,lim),las=1)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
336 grid(col="lightgray", lty="dotted")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
337 lrs = log(cleanrs,10)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
338 hist(lrs,breaks=100,main=paste('After:',maint),xlab="# Reads (log)",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
339 ylab="Count",col="maroon",sub=myTitle,xlim=c(0,lim),las=1)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
340 grid(col="lightgray", lty="dotted")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
341 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
342 par(defpar)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
343 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
344
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
345 cumPlot1 = function(rawrs,cleanrs,maint,myTitle)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
346 { # updated to use ecdf
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
347 pdfname = paste(gsub(" ","", myTitle , fixed=TRUE),"RowsumCum.pdf",sep='_')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
348 pdf(pdfname)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
349 par(mfrow=c(2,1))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
350 lastx = max(rawrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
351 rawe = knots(ecdf(rawrs))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
352 cleane = knots(ecdf(cleanrs))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
353 cy = 1:length(cleane)/length(cleane)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
354 ry = 1:length(rawe)/length(rawe)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
355 plot(rawe,ry,type='l',main=paste('Before',maint),xlab="Log Contig Total Reads",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
356 ylab="Cumulative proportion",col="maroon",log='x',xlim=c(1,lastx),sub=myTitle)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
357 grid(col="blue")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
358 plot(cleane,cy,type='l',main=paste('After',maint),xlab="Log Contig Total Reads",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
359 ylab="Cumulative proportion",col="maroon",log='x',xlim=c(1,lastx),sub=myTitle)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
360 grid(col="blue")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
361 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
362 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
363
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
364
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
365
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
366 doGSEA = function(y=NULL,design=NULL,histgmt="",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
367 bigmt="/data/genomes/gsea/3.1/Abetterchoice_nocgp_c2_c3_c5_symbols_all.gmt",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
368 ntest=0, myTitle="myTitle", outfname="GSEA.xls", minnin=5, maxnin=2000,fdrthresh=0.05,fdrtype="BH")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
369 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
370 sink('Camera.log')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
371 genesets = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
372 if (bigmt > "")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
373 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
374 bigenesets = readLines(bigmt)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
375 genesets = bigenesets
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
376 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
377 if (histgmt > "")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
378 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
379 hgenesets = readLines(histgmt)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
380 if (bigmt > "") {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
381 genesets = rbind(genesets,hgenesets)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
382 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
383 genesets = hgenesets
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
384 } # use only history if no bi
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
385 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
386 print.noquote(paste("@@@read",length(genesets), 'genesets from',histgmt,bigmt))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
387 genesets = strsplit(genesets,'\t') # tabular. genesetid\tURLorwhatever\tgene_1\t..\tgene_n
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
388 outf = outfname
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
389 head=paste(myTitle,'edgeR GSEA')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
390 write(head,file=outfname,append=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
391 ntest=length(genesets)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
392 urownames = toupper(rownames(y))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
393 upcam = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
394 downcam = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
395 for (i in 1:ntest) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
396 gs = unlist(genesets[i])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
397 g = gs[1] # geneset_id
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
398 u = gs[2]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
399 if (u > "") { u = paste("<a href=\'",u,"\'>",u,"</a>",sep="") }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
400 glist = gs[3:length(gs)] # member gene symbols
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
401 glist = toupper(glist)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
402 inglist = urownames %in% glist
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
403 nin = sum(inglist)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
404 if ((nin > minnin) && (nin < maxnin)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
405 ### print(paste('@@found',sum(inglist),'genes in glist'))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
406 camres = camera(y=y,index=inglist,design=design)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
407 if (! is.null(camres)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
408 rownames(camres) = g # gene set name
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
409 camres = cbind(GeneSet=g,URL=u,camres)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
410 if (camres\$Direction == "Up")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
411 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
412 upcam = rbind(upcam,camres) } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
413 downcam = rbind(downcam,camres)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
414 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
415 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
416 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
417 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
418 uscam = upcam[order(upcam\$PValue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
419 unadjp = uscam\$PValue
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
420 uscam\$adjPValue = p.adjust(unadjp,method=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
421 nup = max(10,sum((uscam\$adjPValue < fdrthresh)))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
422 dscam = downcam[order(downcam\$PValue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
423 unadjp = dscam\$PValue
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
424 dscam\$adjPValue = p.adjust(unadjp,method=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
425 ndown = max(10,sum((dscam\$adjPValue < fdrthresh)))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
426 write.table(uscam,file=paste('camera_up',outfname,sep='_'),quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
427 write.table(dscam,file=paste('camera_down',outfname,sep='_'),quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
428 print.noquote(paste('@@@@@ Camera up top',nup,'gene sets:'))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
429 write.table(head(uscam,nup),file="",quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
430 print.noquote(paste('@@@@@ Camera down top',ndown,'gene sets:'))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
431 write.table(head(dscam,ndown),file="",quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
432 sink()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
433 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
434
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
435
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
436
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
437
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
438 doGSEAatonce = function(y=NULL,design=NULL,histgmt="",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
439 bigmt="/data/genomes/gsea/3.1/Abetterchoice_nocgp_c2_c3_c5_symbols_all.gmt",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
440 ntest=0, myTitle="myTitle", outfname="GSEA.xls", minnin=5, maxnin=2000,fdrthresh=0.05,fdrtype="BH")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
441 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
442 sink('Camera.log')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
443 genesets = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
444 if (bigmt > "")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
445 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
446 bigenesets = readLines(bigmt)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
447 genesets = bigenesets
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
448 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
449 if (histgmt > "")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
450 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
451 hgenesets = readLines(histgmt)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
452 if (bigmt > "") {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
453 genesets = rbind(genesets,hgenesets)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
454 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
455 genesets = hgenesets
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
456 } # use only history if no bi
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
457 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
458 print.noquote(paste("@@@read",length(genesets), 'genesets from',histgmt,bigmt))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
459 genesets = strsplit(genesets,'\t') # tabular. genesetid\tURLorwhatever\tgene_1\t..\tgene_n
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
460 outf = outfname
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
461 head=paste(myTitle,'edgeR GSEA')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
462 write(head,file=outfname,append=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
463 ntest=length(genesets)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
464 urownames = toupper(rownames(y))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
465 upcam = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
466 downcam = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
467 incam = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
468 urls = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
469 gsids = c()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
470 for (i in 1:ntest) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
471 gs = unlist(genesets[i])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
472 gsid = gs[1] # geneset_id
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
473 url = gs[2]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
474 if (url > "") { url = paste("<a href=\'",url,"\'>",url,"</a>",sep="") }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
475 glist = gs[3:length(gs)] # member gene symbols
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
476 glist = toupper(glist)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
477 inglist = urownames %in% glist
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
478 nin = sum(inglist)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
479 if ((nin > minnin) && (nin < maxnin)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
480 incam = c(incam,inglist)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
481 gsids = c(gsids,gsid)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
482 urls = c(urls,url)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
483 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
484 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
485 incam = as.list(incam)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
486 names(incam) = gsids
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
487 allcam = camera(y=y,index=incam,design=design)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
488 allcamres = cbind(geneset=gsids,allcam,URL=urls)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
489 for (i in 1:ntest) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
490 camres = allcamres[i]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
491 res = try(test = (camres\$Direction == "Up"))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
492 if ("try-error" %in% class(res)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
493 cat("test failed, camres = :")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
494 print.noquote(camres)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
495 } else { if (camres\$Direction == "Up")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
496 { upcam = rbind(upcam,camres)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
497 } else { downcam = rbind(downcam,camres)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
498 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
499
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
500 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
501 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
502 uscam = upcam[order(upcam\$PValue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
503 unadjp = uscam\$PValue
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
504 uscam\$adjPValue = p.adjust(unadjp,method=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
505 nup = max(10,sum((uscam\$adjPValue < fdrthresh)))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
506 dscam = downcam[order(downcam\$PValue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
507 unadjp = dscam\$PValue
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
508 dscam\$adjPValue = p.adjust(unadjp,method=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
509 ndown = max(10,sum((dscam\$adjPValue < fdrthresh)))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
510 write.table(uscam,file=paste('camera_up',outfname,sep='_'),quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
511 write.table(dscam,file=paste('camera_down',outfname,sep='_'),quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
512 print.noquote(paste('@@@@@ Camera up top',nup,'gene sets:'))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
513 write.table(head(uscam,nup),file="",quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
514 print.noquote(paste('@@@@@ Camera down top',ndown,'gene sets:'))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
515 write.table(head(dscam,ndown),file="",quote=F,sep='\t',row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
516 sink()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
517 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
518
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
519
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
520 edgeIt = function (Count_Matrix=c(),group=c(),out_edgeR=F,out_VOOM=F,out_DESeq2=F,fdrtype='fdr',priordf=5,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
521 fdrthresh=0.05,outputdir='.', myTitle='Differential Counts',libSize=c(),useNDF=F,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
522 filterquantile=0.2, subjects=c(),mydesign=NULL,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
523 doDESeq2=T,doVoom=T,doCamera=T,doedgeR=T,org='hg19',
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
524 histgmt="", bigmt="/data/genomes/gsea/3.1/Abetterchoice_nocgp_c2_c3_c5_symbols_all.gmt",
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
525 doCook=F,DESeq_fitType="parameteric")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
526 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
527 # Error handling
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
528 if (length(unique(group))!=2){
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
529 print("Number of conditions identified in experiment does not equal 2")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
530 q()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
531 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
532 require(edgeR)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
533 options(width = 512)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
534 mt = paste(unlist(strsplit(myTitle,'_')),collapse=" ")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
535 allN = nrow(Count_Matrix)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
536 nscut = round(ncol(Count_Matrix)/2)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
537 colTotmillionreads = colSums(Count_Matrix)/1e6
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
538 counts.dataframe = as.data.frame(c())
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
539 rawrs = rowSums(Count_Matrix)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
540 nonzerod = Count_Matrix[(rawrs > 0),] # remove all zero count genes
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
541 nzN = nrow(nonzerod)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
542 nzrs = rowSums(nonzerod)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
543 zN = allN - nzN
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
544 print('# Quantiles for non-zero row counts:',quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
545 print(quantile(nzrs,probs=seq(0,1,0.1)),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
546 if (useNDF == T)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
547 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
548 gt1rpin3 = rowSums(Count_Matrix/expandAsMatrix(colTotmillionreads,dim(Count_Matrix)) >= 1) >= nscut
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
549 lo = colSums(Count_Matrix[!gt1rpin3,])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
550 workCM = Count_Matrix[gt1rpin3,]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
551 cleanrs = rowSums(workCM)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
552 cleanN = length(cleanrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
553 meth = paste( "After removing",length(lo),"contigs with fewer than ",nscut," sample read counts >= 1 per million, there are",sep="")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
554 print(paste("Read",allN,"contigs. Removed",zN,"contigs with no reads.",meth,cleanN,"contigs"),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
555 maint = paste('Filter >=1/million reads in >=',nscut,'samples')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
556 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
557 useme = (nzrs > quantile(nzrs,filterquantile))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
558 workCM = nonzerod[useme,]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
559 lo = colSums(nonzerod[!useme,])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
560 cleanrs = rowSums(workCM)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
561 cleanN = length(cleanrs)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
562 meth = paste("After filtering at count quantile =",filterquantile,", there are",sep="")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
563 print(paste('Read',allN,"contigs. Removed",zN,"with no reads.",meth,cleanN,"contigs"),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
564 maint = paste('Filter below',filterquantile,'quantile')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
565 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
566 cumPlot(rawrs=rawrs,cleanrs=cleanrs,maint=maint,myTitle=myTitle)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
567 allgenes = rownames(workCM)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
568 reg = "^chr([0-9]+):([0-9]+)-([0-9]+)"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
569 genecards="<a href=\'http://www.genecards.org/index.php?path=/Search/keyword/"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
570 ucsc = paste("<a href=\'http://genome.ucsc.edu/cgi-bin/hgTracks?db=",org,sep='')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
571 testreg = str_match(allgenes,reg)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
572 if (sum(!is.na(testreg[,1]))/length(testreg[,1]) > 0.8) # is ucsc style string
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
573 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
574 print("@@ using ucsc substitution for urls")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
575 contigurls = paste0(ucsc,"&amp;position=chr",testreg[,2],":",testreg[,3],"-",testreg[,4],"\'>",allgenes,"</a>")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
576 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
577 print.noquote("@@ using genecards substitution for urls")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
578 contigurls = paste0(genecards,allgenes,"\'>",allgenes,"</a>")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
579 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
580 print(paste("# Total low count contigs per sample = ",paste(lo,collapse=',')),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
581 cmrowsums = rowSums(workCM)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
582 TName=unique(group)[1]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
583 CName=unique(group)[2]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
584 if (is.null(mydesign)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
585 if (length(subjects) == 0)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
586 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
587 mydesign = model.matrix(~group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
588 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
589 else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
590 subjf = factor(subjects)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
591 mydesign = model.matrix(~subjf+group) # we block on subject so make group last to simplify finding it
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
592 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
593 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
594 print.noquote(paste('Using samples:',paste(colnames(workCM),collapse=',')))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
595 print.noquote('Using design matrix:')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
596 print.noquote(mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
597 if (doedgeR) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
598 sink('edgeR.log')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
599 #### Setup DGEList object
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
600 DGEList = DGEList(counts=workCM, group = group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
601 DGEList = calcNormFactors(DGEList)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
602
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
603 DGEList = estimateGLMCommonDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
604 comdisp = DGEList\$common.dispersion
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
605 DGEList = estimateGLMTrendedDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
606 if (edgeR_priordf > 0) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
607 print.noquote(paste("prior.df =",edgeR_priordf))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
608 DGEList = estimateGLMTagwiseDisp(DGEList,mydesign,prior.df = edgeR_priordf)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
609 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
610 DGEList = estimateGLMTagwiseDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
611 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
612 DGLM = glmFit(DGEList,design=mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
613 DE = glmLRT(DGLM,coef=ncol(DGLM\$design)) # always last one - subject is first if needed
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
614 efflib = DGEList\$samples\$lib.size*DGEList\$samples\$norm.factors
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
615 normData = (1e+06*DGEList\$counts/efflib)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
616 uoutput = cbind(
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
617 Name=as.character(rownames(DGEList\$counts)),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
618 DE\$table,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
619 adj.p.value=p.adjust(DE\$table\$PValue, method=fdrtype),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
620 Dispersion=DGEList\$tagwise.dispersion,totreads=cmrowsums,normData,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
621 DGEList\$counts
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
622 )
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
623 soutput = uoutput[order(DE\$table\$PValue),] # sorted into p value order - for quick toptable
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
624 goodness = gof(DGLM, pcutoff=fdrthresh)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
625 if (sum(goodness\$outlier) > 0) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
626 print.noquote('GLM outliers:')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
627 print(paste(rownames(DGLM)[(goodness\$outlier)],collapse=','),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
628 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
629 print('No GLM fit outlier genes found\n')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
630 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
631 z = limma::zscoreGamma(goodness\$gof.statistic, shape=goodness\$df/2, scale=2)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
632 pdf("edgeR_GoodnessofFit.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
633 qq = qqnorm(z, panel.first=grid(), main="tagwise dispersion")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
634 abline(0,1,lwd=3)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
635 points(qq\$x[goodness\$outlier],qq\$y[goodness\$outlier], pch=16, col="maroon")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
636 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
637 estpriorn = getPriorN(DGEList)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
638 print(paste("Common Dispersion =",comdisp,"CV = ",sqrt(comdisp),"getPriorN = ",estpriorn),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
639 efflib = DGEList\$samples\$lib.size*DGEList\$samples\$norm.factors
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
640 normData = (1e+06*DGEList\$counts/efflib)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
641 uniqueg = unique(group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
642 #### Plot MDS
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
643 sample_colors = match(group,levels(group))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
644 sampleTypes = levels(factor(group))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
645 print.noquote(sampleTypes)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
646 pdf("edgeR_MDSplot.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
647 plotMDS.DGEList(DGEList,main=paste("edgeR MDS for",myTitle),cex=0.5,col=sample_colors,pch=sample_colors)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
648 legend(x="topleft", legend = sampleTypes,col=c(1:length(sampleTypes)), pch=19)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
649 grid(col="blue")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
650 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
651 colnames(normData) = paste( colnames(normData),'N',sep="_")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
652 print(paste('Raw sample read totals',paste(colSums(nonzerod,na.rm=T),collapse=',')))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
653 nzd = data.frame(log(nonzerod + 1e-2,10))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
654 try( boxPlot(rawrs=nzd,cleanrs=log(normData,10),maint='TMM Normalisation',myTitle=myTitle,pdfname="edgeR_raw_norm_counts_box.pdf") )
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
655 write.table(soutput,file=out_edgeR, quote=FALSE, sep="\t",row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
656 tt = cbind(
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
657 Name=as.character(rownames(DGEList\$counts)),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
658 DE\$table,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
659 adj.p.value=p.adjust(DE\$table\$PValue, method=fdrtype),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
660 Dispersion=DGEList\$tagwise.dispersion,totreads=cmrowsums
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
661 )
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
662 print.noquote("# edgeR Top tags\n")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
663 tt = cbind(tt,URL=contigurls) # add to end so table isn't laid out strangely
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
664 tt = tt[order(DE\$table\$PValue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
665 print.noquote(tt[1:50,])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
666 deTags = rownames(uoutput[uoutput\$adj.p.value < fdrthresh,])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
667 nsig = length(deTags)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
668 print(paste('#',nsig,'tags significant at adj p=',fdrthresh),quote=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
669 deColours = ifelse(deTags,'red','black')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
670 pdf("edgeR_BCV_vs_abundance.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
671 plotBCV(DGEList, cex=0.3, main="Biological CV vs abundance")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
672 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
673 dg = DGEList[order(DE\$table\$PValue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
674 #normData = (1e+06 * dg\$counts/expandAsMatrix(dg\$samples\$lib.size, dim(dg)))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
675 efflib = dg\$samples\$lib.size*dg\$samples\$norm.factors
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
676 normData = (1e+06*dg\$counts/efflib)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
677 outpdfname="edgeR_top_100_heatmap.pdf"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
678 hmap2(normData,nsamp=100,TName=TName,group=group,outpdfname=outpdfname,myTitle=paste('edgeR Heatmap',myTitle))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
679 outSmear = "edgeR_smearplot.pdf"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
680 outMain = paste("Smear Plot for ",TName,' Vs ',CName,' (FDR@',fdrthresh,' N = ',nsig,')',sep='')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
681 smearPlot(DGEList=DGEList,deTags=deTags, outSmear=outSmear, outMain = outMain)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
682 qqPlot(descr=paste(myTitle,'edgeR adj p QQ plot'),pvector=tt\$adj.p.value,outpdf='edgeR_qqplot.pdf')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
683 norm.factor = DGEList\$samples\$norm.factors
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
684 topresults.edgeR = soutput[which(soutput\$adj.p.value < fdrthresh), ]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
685 edgeRcountsindex = which(allgenes %in% rownames(topresults.edgeR))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
686 edgeRcounts = rep(0, length(allgenes))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
687 edgeRcounts[edgeRcountsindex] = 1 # Create venn diagram of hits
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
688 sink()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
689 } ### doedgeR
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
690 if (doDESeq2 == T)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
691 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
692 sink("DESeq2.log")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
693 # DESeq2
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
694 require('DESeq2')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
695 library('RColorBrewer')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
696 if (length(subjects) == 0)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
697 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
698 pdata = data.frame(Name=colnames(workCM),Rx=group,row.names=colnames(workCM))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
699 deSEQds = DESeqDataSetFromMatrix(countData = workCM, colData = pdata, design = formula(~ Rx))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
700 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
701 pdata = data.frame(Name=colnames(workCM),Rx=group,subjects=subjects,row.names=colnames(workCM))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
702 deSEQds = DESeqDataSetFromMatrix(countData = workCM, colData = pdata, design = formula(~ subjects + Rx))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
703 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
704 #DESeq2 = DESeq(deSEQds,fitType='local',pAdjustMethod=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
705 #rDESeq = results(DESeq2)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
706 #newCountDataSet(workCM, group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
707 deSeqDatsizefac = estimateSizeFactors(deSEQds)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
708 deSeqDatdisp = estimateDispersions(deSeqDatsizefac,fitType=DESeq_fitType)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
709 resDESeq = nbinomWaldTest(deSeqDatdisp, pAdjustMethod=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
710 rDESeq = as.data.frame(results(resDESeq))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
711 rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
712 srDESeq = rDESeq[order(rDESeq\$pvalue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
713 write.table(srDESeq,file=out_DESeq2, quote=FALSE, sep="\t",row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
714 qqPlot(descr=paste(myTitle,'DESeq2 adj p qq plot'),pvector=rDESeq\$padj,outpdf='DESeq2_qqplot.pdf')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
715 cat("# DESeq top 50\n")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
716 rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums,URL=contigurls)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
717 srDESeq = rDESeq[order(rDESeq\$pvalue),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
718 print.noquote(srDESeq[1:50,])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
719 topresults.DESeq = rDESeq[which(rDESeq\$padj < fdrthresh), ]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
720 DESeqcountsindex = which(allgenes %in% rownames(topresults.DESeq))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
721 DESeqcounts = rep(0, length(allgenes))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
722 DESeqcounts[DESeqcountsindex] = 1
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
723 pdf("DESeq2_dispersion_estimates.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
724 plotDispEsts(resDESeq)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
725 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
726 ysmall = abs(min(rDESeq\$log2FoldChange))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
727 ybig = abs(max(rDESeq\$log2FoldChange))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
728 ylimit = min(4,ysmall,ybig)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
729 pdf("DESeq2_MA_plot.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
730 plotMA(resDESeq,main=paste(myTitle,"DESeq2 MA plot"),ylim=c(-ylimit,ylimit))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
731 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
732 rlogres = rlogTransformation(resDESeq)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
733 sampledists = dist( t( assay(rlogres) ) )
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
734 sdmat = as.matrix(sampledists)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
735 pdf("DESeq2_sample_distance_plot.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
736 heatmap.2(sdmat,trace="none",main=paste(myTitle,"DESeq2 sample distances"),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
737 col = colorRampPalette( rev(brewer.pal(9, "RdBu")) )(255))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
738 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
739 ###outpdfname="DESeq2_top50_heatmap.pdf"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
740 ###hmap2(sresDESeq,nsamp=50,TName=TName,group=group,outpdfname=outpdfname,myTitle=paste('DESeq2 vst rlog Heatmap',myTitle))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
741 sink()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
742 result = try( (ppca = plotPCA( varianceStabilizingTransformation(deSeqDatdisp,blind=T), intgroup=c("Rx","Name")) ) )
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
743 if ("try-error" %in% class(result)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
744 print.noquote('DESeq2 plotPCA failed.')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
745 } else {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
746 pdf("DESeq2_PCA_plot.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
747 #### wtf - print? Seems needed to get this to work
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
748 print(ppca)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
749 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
750 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
751 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
752
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
753 if (doVoom == T) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
754 sink('Voom.log')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
755 if (doedgeR == F) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
756 #### Setup DGEList object
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
757 DGEList = DGEList(counts=workCM, group = group)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
758 DGEList = calcNormFactors(DGEList)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
759 DGEList = estimateGLMCommonDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
760 DGEList = estimateGLMTrendedDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
761 DGEList = estimateGLMTagwiseDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
762 DGEList = estimateGLMTagwiseDisp(DGEList,mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
763 norm.factor = DGEList\$samples\$norm.factors
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
764 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
765 pdf("Voom_mean_variance_plot.pdf")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
766 dat.voomed = voom(DGEList, mydesign, plot = TRUE, lib.size = colSums(workCM) * norm.factor)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
767 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
768 # Use limma to fit data
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
769 fit = lmFit(dat.voomed, mydesign)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
770 fit = eBayes(fit)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
771 rvoom = topTable(fit, coef = length(colnames(mydesign)), adj = fdrtype, n = Inf, sort="none")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
772 qqPlot(descr=paste(myTitle,'Voom-limma adj p QQ plot'),pvector=rvoom\$adj.P.Val,outpdf='Voom_qqplot.pdf')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
773 rownames(rvoom) = rownames(workCM)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
774 rvoom = cbind(rvoom,NReads=cmrowsums)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
775 srvoom = rvoom[order(rvoom\$P.Value),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
776 write.table(srvoom,file=out_VOOM, quote=FALSE, sep="\t",row.names=F)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
777 rvoom = cbind(rvoom,NReads=cmrowsums,URL=contigurls)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
778 srvoom = rvoom[order(rvoom\$P.Value),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
779 cat("# Voom top 50\n")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
780 print(srvoom[1:50,])
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
781 # Use an FDR cutoff to find interesting samples for edgeR, DESeq and voom/limma
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
782 topresults.voom = rvoom[which(rvoom\$adj.P.Val < fdrthresh), ]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
783 voomcountsindex = which(allgenes %in% topresults.voom\$ID)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
784 voomcounts = rep(0, length(allgenes))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
785 voomcounts[voomcountsindex] = 1
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
786 sink()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
787 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
788
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
789 if (doCamera) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
790 doGSEA(y=DGEList,design=mydesign,histgmt=histgmt,bigmt=bigmt,ntest=20,myTitle=myTitle,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
791 outfname=paste(mt,"GSEA.xls",sep="_"),fdrthresh=fdrthresh,fdrtype=fdrtype)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
792 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
793
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
794 if ((doDESeq2==T) || (doVoom==T) || (doedgeR==T)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
795 if ((doVoom==T) && (doDESeq2==T) && (doedgeR==T)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
796 vennmain = paste(mt,'Voom,edgeR and DESeq2 overlap at FDR=',fdrthresh)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
797 counts.dataframe = data.frame(edgeR = edgeRcounts, DESeq2 = DESeqcounts,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
798 VOOM_limma = voomcounts, row.names = allgenes)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
799 } else if ((doDESeq2==T) && (doedgeR==T)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
800 vennmain = paste(mt,'DESeq2 and edgeR overlap at FDR=',fdrthresh)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
801 counts.dataframe = data.frame(edgeR = edgeRcounts, DESeq2 = DESeqcounts, row.names = allgenes)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
802 } else if ((doVoom==T) && (doedgeR==T)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
803 vennmain = paste(mt,'Voom and edgeR overlap at FDR=',fdrthresh)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
804 counts.dataframe = data.frame(edgeR = edgeRcounts, VOOM_limma = voomcounts, row.names = allgenes)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
805 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
806
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
807 if (nrow(counts.dataframe > 1)) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
808 counts.venn = vennCounts(counts.dataframe)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
809 vennf = "Venn_significant_genes_overlap.pdf"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
810 pdf(vennf)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
811 vennDiagram(counts.venn,main=vennmain,col="maroon")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
812 dev.off()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
813 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
814 } #### doDESeq2 or doVoom
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
815
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
816 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
817 #### Done
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
818
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
819 ###sink(stdout(),append=T,type="message")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
820 builtin_gmt = ""
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
821 history_gmt = ""
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
822 history_gmt_name = ""
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
823 out_edgeR = F
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
824 out_DESeq2 = F
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
825 out_VOOM = "$out_VOOM"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
826 doDESeq2 = $DESeq2.doDESeq2 # make these T or F
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
827 doVoom = $doVoom
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
828 doCamera = F
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
829 doedgeR = $edgeR.doedgeR
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
830 edgeR_priordf = 0
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
831
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
832
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
833 #if $doVoom == "T":
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
834 out_VOOM = "$out_VOOM"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
835 #end if
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
836
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
837 #if $DESeq2.doDESeq2 == "T":
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
838 out_DESeq2 = "$out_DESeq2"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
839 DESeq_fitType = "$DESeq2.DESeq_fitType"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
840 #end if
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
841
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
842 #if $edgeR.doedgeR == "T":
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
843 out_edgeR = "$out_edgeR"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
844 edgeR_priordf = $edgeR.edgeR_priordf
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
845 #end if
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
846
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
847
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
848 if (sum(c(doedgeR,doVoom,doDESeq2)) == 0)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
849 {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
850 write("No methods chosen - nothing to do! Please try again after choosing one or more methods", stderr())
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
851 quit(save="no",status=2)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
852 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
853
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
854 Out_Dir = "$html_file.files_path"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
855 Input = "$input1"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
856 TreatmentName = "$treatment_name"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
857 TreatmentCols = "$Treat_cols"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
858 ControlName = "$control_name"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
859 ControlCols= "$Control_cols"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
860 org = "$input1.dbkey"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
861 if (org == "") { org = "hg19"}
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
862 fdrtype = "$fdrtype"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
863 fdrthresh = $fdrthresh
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
864 useNDF = $useNDF
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
865 fQ = $fQ # non-differential centile cutoff
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
866 myTitle = "$title"
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
867 sids = strsplit("$subjectids",',')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
868 subjects = unlist(sids)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
869 nsubj = length(subjects)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
870 TCols = as.numeric(strsplit(TreatmentCols,",")[[1]])-1
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
871 CCols = as.numeric(strsplit(ControlCols,",")[[1]])-1
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
872 cat('Got TCols=')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
873 cat(TCols)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
874 cat('; CCols=')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
875 cat(CCols)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
876 cat('\n')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
877 useCols = c(TCols,CCols)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
878 if (file.exists(Out_Dir) == F) dir.create(Out_Dir)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
879 Count_Matrix = read.table(Input,header=T,row.names=1,sep='\t') #Load tab file assume header
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
880 snames = colnames(Count_Matrix)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
881 nsamples = length(snames)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
882 if (nsubj > 0 & nsubj != nsamples) {
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
883 options("show.error.messages"=T)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
884 mess = paste('Fatal error: Supplied subject id list',paste(subjects,collapse=','),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
885 'has length',nsubj,'but there are',nsamples,'samples',paste(snames,collapse=','))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
886 write(mess, stderr())
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
887 quit(save="no",status=4)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
888 }
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
889 if (length(subjects) != 0) {subjects = subjects[useCols]}
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
890 Count_Matrix = Count_Matrix[,useCols] ### reorder columns
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
891 rn = rownames(Count_Matrix)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
892 islib = rn %in% c('librarySize','NotInBedRegions')
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
893 LibSizes = Count_Matrix[subset(rn,islib),][1] # take first
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
894 Count_Matrix = Count_Matrix[subset(rn,! islib),]
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
895 group = c(rep(TreatmentName,length(TCols)), rep(ControlName,length(CCols)) ) #Build a group descriptor
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
896 group = factor(group, levels=c(ControlName,TreatmentName))
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
897 colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_") #Relable columns
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
898 results = edgeIt(Count_Matrix=Count_Matrix,group=group, out_edgeR=out_edgeR, out_VOOM=out_VOOM, out_DESeq2=out_DESeq2,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
899 fdrtype='BH',mydesign=NULL,priordf=edgeR_priordf,fdrthresh=fdrthresh,outputdir='.',
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
900 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=subjects,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
901 doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
902 histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
903 sessionInfo()
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
904 ]]>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
905 </configfile>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
906 </configfiles>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
907 <help>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
908
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
909 **What it does**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
910
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
911 Allows short read sequence counts from controlled experiments to be analysed for differentially expressed genes.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
912 Optionally adds a term for subject if not all samples are independent or if some other factor needs to be blocked in the design.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
913
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
914 **Input**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
915
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
916 Requires a count matrix as a tabular file. These are best made using the companion HTSeq_ based counter Galaxy wrapper
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
917 and your fave gene model to generate inputs. Each row is a genomic feature (gene or exon eg) and each column the
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
918 non-negative integer count of reads from one sample overlapping the feature.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
919 The matrix must have a header row uniquely identifying the source samples, and unique row names in
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
920 the first column. Typically the row names are gene symbols or probe ids for downstream use in GSEA and other methods.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
921
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
922 **Specifying comparisons**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
923
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
924 This is basically dumbed down for two factors - case vs control.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
925
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
926 More complex interfaces are possible but painful at present.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
927 Probably need to specify a phenotype file to do this better.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
928 Work in progress. Send code.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
929
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
930 If you have (eg) paired samples and wish to include a term in the GLM to account for some other factor (subject in the case of paired samples),
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
931 put a comma separated list of indicators for every sample (whether modelled or not!) indicating (eg) the subject number or
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
932 A list of integers, one for each subject or an empty string if samples are all independent.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
933 If not empty, there must be exactly as many integers in the supplied integer list as there are columns (samples) in the count matrix.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
934 Integers for samples that are not in the analysis *must* be present in the string as filler even if not used.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
935
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
936 So if you have 2 pairs out of 6 samples, you need to put in unique integers for the unpaired ones
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
937 eg if you had 6 samples with the first two independent but the second and third pairs each being from independent subjects. you might use
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
938 8,9,1,1,2,2
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
939 as subject IDs to indicate two paired samples from the same subject in columns 3/4 and 5/6
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
940
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
941 **Methods available**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
942
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
943 You can run 3 popular Bioconductor packages available for count data.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
944
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
945 edgeR - see edgeR_ for details
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
946
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
947 VOOM/limma - see limma_VOOM_ for details
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
948
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
949 DESeq2 - see DESeq2_ for details
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
950
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
951 and optionally camera in edgeR which works better if MSigDB is installed.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
952
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
953 **Outputs**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
954
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
955 Some helpful plots and analysis results. Note that most of these are produced using R code
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
956 suggested by the excellent documentation and vignettes for the Bioconductor
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
957 packages invoked. The Tool Factory is used to automatically lay these out for you to enjoy.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
958
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
959 **Note on Voom**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
960
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
961 The voom from limma version 3.16.6 help in R includes this from the authors - but you should read the paper to interpret this method.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
962
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
963 This function is intended to process RNA-Seq or ChIP-Seq data prior to linear modelling in limma.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
964
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
965 voom is an acronym for mean-variance modelling at the observational level.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
966 The key concern is to estimate the mean-variance relationship in the data, then use this to compute appropriate weights for each observation.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
967 Count data almost show non-trivial mean-variance relationships. Raw counts show increasing variance with increasing count size, while log-counts typically show a decreasing mean-variance trend.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
968 This function estimates the mean-variance trend for log-counts, then assigns a weight to each observation based on its predicted variance.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
969 The weights are then used in the linear modelling process to adjust for heteroscedasticity.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
970
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
971 In an experiment, a count value is observed for each tag in each sample. A tag-wise mean-variance trend is computed using lowess.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
972 The tag-wise mean is the mean log2 count with an offset of 0.5, across samples for a given tag.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
973 The tag-wise variance is the quarter-root-variance of normalized log2 counts per million values with an offset of 0.5, across samples for a given tag.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
974 Tags with zero counts across all samples are not included in the lowess fit. Optional normalization is performed using normalizeBetweenArrays.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
975 Using fitted values of log2 counts from a linear model fit by lmFit, variances from the mean-variance trend were interpolated for each observation.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
976 This was carried out by approxfun. Inverse variance weights can be used to correct for mean-variance trend in the count data.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
977
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
978
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
979 Author(s)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
980
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
981 Charity Law and Gordon Smyth
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
982
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
983 References
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
984
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
985 Law, CW (2013). Precision weights for gene expression analysis. PhD Thesis. University of Melbourne, Australia.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
986
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
987 Law, CW, Chen, Y, Shi, W, Smyth, GK (2013). Voom! Precision weights unlock linear model analysis tools for RNA-seq read counts.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
988 Technical Report 1 May 2013, Bioinformatics Division, Walter and Eliza Hall Institute of Medical Reseach, Melbourne, Australia.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
989 http://www.statsci.org/smyth/pubs/VoomPreprint.pdf
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
990
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
991 See Also
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
992
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
993 A voom case study is given in the edgeR User's Guide.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
994
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
995 vooma is a similar function but for microarrays instead of RNA-seq.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
996
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
997
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
998 ***old rant on changes to Bioconductor package variable names between versions***
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
999
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1000 The edgeR authors made a small cosmetic change in the name of one important variable (from p.value to PValue)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1001 breaking this and all other code that assumed the old name for this variable,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1002 between edgeR2.4.4 and 2.4.6 (the version for R 2.14 as at the time of writing).
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1003 This means that all code using edgeR is sensitive to the version. I think this was a very unwise thing
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1004 to do because it wasted hours of my time to track down and will similarly cost other edgeR users dearly
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1005 when their old scripts break. This tool currently now works with 2.4.6.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1006
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1007 **Note on prior.N**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1008
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1009 http://seqanswers.com/forums/showthread.php?t=5591 says:
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1010
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1011 *prior.n*
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1012
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1013 The value for prior.n determines the amount of smoothing of tagwise dispersions towards the common dispersion.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1014 You can think of it as like a "weight" for the common value. (It is actually the weight for the common likelihood
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1015 in the weighted likelihood equation). The larger the value for prior.n, the more smoothing, i.e. the closer your
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1016 tagwise dispersion estimates will be to the common dispersion. If you use a prior.n of 1, then that gives the
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1017 common likelihood the weight of one observation.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1018
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1019 In answer to your question, it is a good thing to squeeze the tagwise dispersions towards a common value,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1020 or else you will be using very unreliable estimates of the dispersion. I would not recommend using the value that
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1021 you obtained from estimateSmoothing()---this is far too small and would result in virtually no moderation
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1022 (squeezing) of the tagwise dispersions. How many samples do you have in your experiment?
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1023 What is the experimental design? If you have few samples (less than 6) then I would suggest a prior.n of at least 10.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1024 If you have more samples, then the tagwise dispersion estimates will be more reliable,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1025 so you could consider using a smaller prior.n, although I would hesitate to use a prior.n less than 5.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1026
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1027
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1028 From Bioconductor Digest, Vol 118, Issue 5, Gordon writes:
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1029
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1030 Dear Dorota,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1031
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1032 The important settings are prior.df and trend.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1033
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1034 prior.n and prior.df are related through prior.df = prior.n * residual.df,
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1035 and your experiment has residual.df = 36 - 12 = 24. So the old setting of
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1036 prior.n=10 is equivalent for your data to prior.df = 240, a very large
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1037 value. Going the other way, the new setting of prior.df=10 is equivalent
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1038 to prior.n=10/24.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1039
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1040 To recover old results with the current software you would use
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1041
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1042 estimateTagwiseDisp(object, prior.df=240, trend="none")
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1043
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1044 To get the new default from old software you would use
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1045
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1046 estimateTagwiseDisp(object, prior.n=10/24, trend=TRUE)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1047
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1048 Actually the old trend method is equivalent to trend="loess" in the new
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1049 software. You should use plotBCV(object) to see whether a trend is
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1050 required.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1051
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1052 Note you could also use
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1053
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1054 prior.n = getPriorN(object, prior.df=10)
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1055
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1056 to map between prior.df and prior.n.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1057
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1058 ----
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1059
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1060 **Attributions**
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1061
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1062 edgeR - edgeR_
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1063
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1064 VOOM/limma - limma_VOOM_
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1065
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1066 DESeq2 - DESeq2_ for details
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1067
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1068 See above for Bioconductor package documentation for packages exposed in Galaxy by this tool and app store package.
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1069
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1070 Galaxy_ (that's what you are using right now!) for gluing everything together
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1071
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1072 Otherwise, all code and documentation comprising this tool was written by Ross Lazarus and is
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1073 licensed to you under the LGPL_ like other rgenetics artefacts
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1074
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1075 .. _LGPL: http://www.gnu.org/copyleft/lesser.html
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1076 .. _HTSeq: http://www-huber.embl.de/users/anders/HTSeq/doc/index.html
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1077 .. _edgeR: http://www.bioconductor.org/packages/release/bioc/html/edgeR.html
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1078 .. _DESeq2: http://www.bioconductor.org/packages/release/bioc/html/DESeq2.html
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1079 .. _limma_VOOM: http://www.bioconductor.org/packages/release/bioc/html/limma.html
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1080 .. _Galaxy: http://getgalaxy.org
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1081 </help>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1082
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1083 </tool>
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1084
8051d14cf819 update with better filtering
fubar
parents:
diff changeset
1085