source: plugins/base2/net.sf.basedb.normalizers/trunk/META-INF/extensions.xml @ 1454

Last change on this file since 1454 was 1454, checked in by Nicklas Nordborg, 11 years ago

Preparing 'Normalization package 1.1-beta' release.

  • Property svn:eol-style set to native
  • Property svn:keywords set to Date Id
File size: 4.8 KB
1<?xml version="1.0" encoding="UTF-8" ?>
2<extensions xmlns="">
3  <about>
4    <name>Normalization plug-ins package</name>
5    <description>
6      This package is a compilation of normalisers for expression data.
7      Common to most of the plug-ins is that they work on bioassay sets
8      with either 1-channel or 2-channel data. The algorithms are working
9      on expression values, that is for 2-channel data, ratio ch1/ch2 are
10      used.
11    </description>
12    <version>1.1-beta</version>
13    <min-base-version>3.0.0</min-base-version>
14    <copyright>BASE development team</copyright>
15    <email></email>
16    <url></url>
17  </about>
19  <plugin-definition id="AverageNormalization">
20    <about>
21      <name>Average normalization</name>
22      <description>
23        This plug-in scales the expression values for an assay with a
24        factor, "S", equal to the ratio of either:
26        i) the geometric mean of the expression values of all spots in the
27        bioassay set divided by the assay average, or
29        ii) a user defined value divided by the assay average.
31        The new expression values will become "S" times the original
32        expression value. Background subtraction and proper filtration
33        have to be done before running this plug-in.
35        This plug-in supports 1-channel and 2-channel data.
36      </description>     
37    </about>   
38    <plugin-class>net.sf.basedb.plugins.AverageNormalization</plugin-class>
39    <settings>
40      <property name="everyone-use">1</property>
41    </settings>
42  </plugin-definition>
44  <plugin-definition id="QuantileNormalization">
45    <about>
46      <name>Quantile normalization</name>
47      <description>
48        In quantile normalization each assay data is sorted in ascending
49        expression value order and added to a matrix as columns. The matrix
50        rows will contain mixed probes (also known as reporters or genes)
51        decided by their rank. For each row in the matrix, the expression
52        values are replaced with the row average value. Finally, each assay
53        is reordered into its original order to retain a standard
54        expression matrix were each row represents one probe. Assays are
55        not mixed.
57        Background subtraction and proper filtration should be done on the
58        bioassay set before running this plug-in. The bioassay set must not
59        contain any missing values.
61        This plug-in supports 1-channel and 2-channel data.
62      </description>     
63    </about>   
64    <plugin-class>net.sf.basedb.plugins.QuantileNormalization</plugin-class>
65    <settings>
66      <property name="everyone-use">1</property>
67    </settings>
68  </plugin-definition>
70  <plugin-definition id="qQuantileNormalization">
71    <about>
72      <name>qQuantile normalization</name>
73      <description>
74        The qQuantile normalization is inspired by the 'Cubic Spline' normalization
75        in Illumina Beadstudio and the work by Workman et al.,
78        In qQuantile normalization, all assays (including the target) are sorted in
79        increasing intensity. The sorted list of probe intensities are partitioned
80        into q groups, and each of theses q groups are adjusted (normalized) with the
81        corresponding target group. After normalization the intensity distribution of
82        each assay will be approximately the same as the target distribution. q is
83        calculated as q=max(10,min(100,target_size/10)). The program will stop if the
84        number of well defined expression values in the target or any of the assays
85        in the set is smaller than q.
87        The target is defined by selecting a subset of the assays in the bioassay set,
88        and the target expression values are the medians of probe intensities over the
89        bioassay set. Probes with no well defined measurements in the bioassay set are
90        simply ignored in target calculation.
92        Since the normalization calculations are based on geometric means and performed
93        in log space the intensities must be positive and larger than 0. Rather than
94        expecting the user of qQuantile normalization to remove such intensity the
95        underlying algorithm silently ignores zero and negative intensities.
97        Background subtraction and proper filtration should be done on the bioassay set
98        before running this plug-in.
100        Only 1-channel data is supported.
101      </description>     
102    </about>
103    <plugin-class>net.sf.basedb.plugins.qQuantileNormalization</plugin-class>
104    <settings>
105      <property name="deprecated">1</property>
106    </settings>
107  </plugin-definition>
109  <plugin-definition id="RankInvariantNormalization">
110    <about>
111      <name>Rank invariant normalization</name>
112      <description>
113        The development of this plug-in is still in progress
114      </description>     
115    </about>   
116    <plugin-class>net.sf.basedb.plugins.RankInvariantNormalization</plugin-class>
117  </plugin-definition>
Note: See TracBrowser for help on using the repository browser.