source: plugins/base2/net.sf.basedb.illumina/trunk/README_PluginDetails @ 1218

Last change on this file since 1218 was 1218, checked in by Jari Häkkinen, 12 years ago

Addresses #231. Improved documents and display link to illumina package web page when error occurs.

  • Property svn:eol-style set to native
  • Property svn:keywords set to Id
  • Property svn:mime-type set to text/x-trac-wiki; charset=utf-8
File size: 5.7 KB
RevLine 
[1038]1$Id: README_PluginDetails 1218 2010-03-29 08:48:15Z jari $
2
3== Introduction ==
4
5There are many plug-ins in the Illumina plug-in package for BASE. This
6file gives detailed information on some of the contributed plug-ins.
7
8
[1218]9
[1203]10== Illumina expression background correction plug-in ==
11
12This plug-in will remove a per-slide global background from all
13spots. The background is calculated from a set of negative control
14spot on the array. See implementation details below for details.
15
16=== Parameters ===
17
[1207]18There is one parameter to set that specifies how background
19intensities should be calculated. Allowed values are median or mean,
20i.e. the background is either the median or the mean of the negative
21control spots on the array.
[1203]22
[1207]23The expression of the background probes is optionally saved to a
24file. The default is not to save the expression matrix but this can be
25changed during job configuration.
26
[1203]27=== Implementation details ===
28
29Each assay is treated separately, i.e., no samples are combined
30together. All calculations are made on the current bioassay data
31implying that this plug-in should be used early in analysis and before
32background spots are removed.
33
34A spot is considered to be a negative control spot if it has an
35''Control group name'' exactly matching the string ''negative''.
36
37
38
[1038]39== Illumina detection P-value calculation ==
40
41This plug-in implements !BeadStudio like detection P-value calculations
42for Illumina expression data (see
[1113]43http://www.genomecenter.ucdavis.edu/expression_analysis/documents/illumina_normalization_081201.pdf)
[1038]44on detection P-values.
45
46The plug-in will ''always'' base the detection P-value calculation on
47raw data values, ''i.e.'', the mean raw intensity for the different
[1168]48signals. By default the calculations are based on negative controls
49available in the root bioassay set for the current analysis
50branch. The user may change this to only use negative controls in the
51current bioassay set.
[1038]52
[1113]53The detection P-value plug-in does not filter the assays, it provides
54the detection P-values usable in a filter step after running this
55plug-in.
[1038]56
[1113]57=== Parameters ===
58
[1122]59The plug-in requires input of array type since detection P-value are
60calculated differently depending on array type.
[1113]61
[1168]62Users may select to use negative controls in the current bioassay set
63only. The default behaviour is to use all negative controls in the
64root bioassay set for the current bioassay set.
65
66A cut off parameter is available to exclude outliers within
[1113]67the negative controls. The `cutoff` defines the acceptable negative
68control signal range
69{{{
70median-MAD*cutoff < I < median+MAD*cutoff
71}}}
72where `MAD` is the median absolute deviation.
73
74=== Implementation details ===
75
76Each assay is treated separately, i.e., no samples are combined
[1122]77together. All calculations are made on the raw bead-type level data,
78i.e., on the average expression value for each bead type and raw data
79is always used irrespective when in analysis the detection P-value is
80calculated.
[1113]81
[1167]82''Pvalue calculation for whole genome arrays:''
[1113]83
84For all signals `i` calculate the detection P-value as
[1122]85`Pvalue = 1-R/N` where `R` is the rank of the signal `i` relative to the
[1114]86negative controls and `N` is the number of negative controls.
[1113]87
[1167]88''Pvalue calculation for others array types (DASL, miRNA, !VeraCode
89DASL, and Focused Arrays):''
[1113]90
[1114]91For all signals `i` calculate the detection P-value as
[1167]92`Pvalue = 1/2 - 1/2 * erf( [i-AvgControl]/StdControl/sqrt(2) )` where
[1114]93`AvgControl` is the average intensity of the negative controls,
[1122]94`StdControl` is the standard deviation of the the negative controls,
95and `erf` is the error function
[1114]96(http://mathworld.wolfram.com/Erf.html). The error function is used
[1122]97for arguments within the range (-4,4). To save CPU cycles, the function
[1114]98value for arguments outside this range is set to -1 and 1,
99respectively.
100
[1218]101A spot is considered to be a negative control spot if it has an
102''Control group name'' exactly matching the string ''negative''.
[1114]103
[1203]104
[1218]105
[1179]106== Control Summary plots ==
107
[1197]108This extension provides overview plots for Illumina expression data. The
[1179]109'Overview plots' tab becomes available in the experiment analysis tree
110when the user selects a bioassay set. The plots are automatically
111generated when the user clicks on the Overview plots tab. The display
112of the plots cannot be changed by the user and the same plot is shown
113irrespective which bioassay set is selected.
114
115Currently two control summary curves are generated in one plot; The
116average intensity of perfect match beads in each assay, and the
117average intensity of housekeeping beads in each assay.
118
119The average intensity I_avg is calculated as
120
121    I_avg = sum[Iraw_i] / N
122
123where N is the number of bead types in the sum, Iraw_i is the raw
124mean intensity for bead type i.
125
126A bead type is considered to belong to the perfect match group if it
127is annotated with ':pm' in the reporter annotation column '[Rep]
128Control group id', and a bead type is grouped as housekeeping if it is
129annotated with 'housekeeping' in reporter annotator column '[Rep]
130Control group id'.
131
[1038]132----------------------------------------------------------------------
133{{{
[1203]134Copyright (C) 2009, 2010 Jari Häkkinen
[1038]135
136This file is part of Illumina plug-in package for BASE.
137Available at http://baseplugins.thep.lu.se/
138BASE main site: http://base.thep.lu.se/
139
140This is free software; you can redistribute it and/or
141modify it under the terms of the GNU General Public License
142as published by the Free Software Foundation; either version 3
143of the License, or (at your option) any later version.
144
145The software is distributed in the hope that it will be useful,
146but WITHOUT ANY WARRANTY; without even the implied warranty of
147MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
148GNU General Public License for more details.
149
150You should have received a copy of the GNU General Public License
151along with BASE. If not, see <http://www.gnu.org/licenses/>.
152}}}
153----------------------------------------------------------------------
Note: See TracBrowser for help on using the repository browser.