Changeset 1064

Timestamp:

May 14, 2009, 12:26:03 PM (14 years ago)

Author:

Jari Häkkinen

Message:

Addresses #118. Calculate target with median instead of mean.

Location:

plugins/base2/net.sf.basedb.normalizers/trunk

Files:

• README (modified) (1 diff)
• src/net/sf/basedb/plugins/qQuantileNormalization.java (modified) (2 diffs)

plugins/base2/net.sf.basedb.normalizers/trunk/README

@@ -50 +50 @@
 == qQuantile normalization ==
 
-Write me.
+qQN geometric mean non-logged values!
+qQN must have positive numbers!
+
+The assays are normed against a selectable sub-set of the assays
+... If a probe has no well-defined measurement (i.e., no assay in
+the reference has a well defined value for a probe) it is simply
+ignored from the target distribution.
+
+Ja, jag tankte att man far justera q till maximalt antal icke nan ...
+men som sagt an sa lange ar q==100.
+
+Den begränsning som finns nu är att det måste finnas lagom många
+väldefinierade mätvärden per assay (några hundra per assay) annars
+kraschar nog programmet. Jag kommer att lösa detta genom att välja
+antalet bins i distributionsberäkningen som #bins=max(100,N_i/10)
+i=1...#assays och kräva att #bins är minst 10. Kravet är alltså att
+varje assay måste ha minst 100 väldefinierade punkter. Om inte kravet
+är uppfyllt stannar programmet med ett någorlunda trevligt meddelande.
+
+In q-quantile normalization each assay data is sorted in ascending
+expression value order and added to a matrix as columns. The matrix
+rows will contain mixed probes (also known as reporters or genes)
+decided by their rank. For each row in the matrix, the expression
+values are replaced with the row average value. Finally, each assay is
+reordered into its original order to retain a standard expression
+matrix were each row represents one probe. Assays are not mixed.
+
+Background subtraction and proper filtration should be done on the
+bioassay set before running this plug-in. The bioassay set must not
+contain any missing values.
 
 The qQuantile normalization is inspired by the 'Cubic Spline'

plugins/base2/net.sf.basedb.normalizers/trunk/src/net/sf/basedb/plugins/qQuantileNormalization.java

@@ -81 +81 @@
         "Background subtraction and proper filtration have to be done on the " +
         "bioassay set before running this plug-in.\n\n" +
-        "1-channel or 2-channel are supported\n\n" +
+        "Only 1-channel data is supported\n\n" +
         "This plug-in is part of the BASE normalizers package.",
         Normalizations.getVersion(),
@@ -215 +215 @@
                      "--assay-data", assaydata,
                      "--in-data", data_before,
-                     "--out-data", data_after };
+                     "--out-data", data_after,
+                     "--median" };
     execute(cmd);
   }