Difference between revisions of "Web site v3.87 Release Notes"

From GeneSetEnrichmentAnalysisWiki
Jump to navigation Jump to search
m
m
Line 1: Line 1:
 
<a href="http://www.broadinstitute.org/gsea/">GSEA Home</a> |  <a href="http://www.broadinstitute.org/gsea/downloads.jsp">Downloads</a>  | <a href="http://www.broadinstitute.org/gsea/msigdb/">Molecular  Signatures Database</a> | <a href="http://www.broadinstitute.org/cancer/software/gsea/wiki/index.php/Main_Page">Documentation</a> | <a href="http://www.broadinstitute.org/gsea/contact.jsp">Contact</a>  <br />
 
<a href="http://www.broadinstitute.org/gsea/">GSEA Home</a> |  <a href="http://www.broadinstitute.org/gsea/downloads.jsp">Downloads</a>  | <a href="http://www.broadinstitute.org/gsea/msigdb/">Molecular  Signatures Database</a> | <a href="http://www.broadinstitute.org/cancer/software/gsea/wiki/index.php/Main_Page">Documentation</a> | <a href="http://www.broadinstitute.org/gsea/contact.jsp">Contact</a>  <br />
 
<br />
 
<br />
Release 3.87 of the GSEA public Web site contains several updates and bug fixes, including:<br />
+
Release 3.87 of the GSEA public Web site contains several updates and bug fixes, including:
 
<ul>
 
<ul>
     <li>The Web site now exposes all gene set original members, in addition to the corresponding Entrez Gene IDs and gene symbols; unmapped members are indicated with a grey background and titleAdding FDR q-values to the Compute Overlaps output: Compute Overlaps now reports both raw hypergeometric p-values and their analogs after multiple test corrections according to [http://en.wikipedia.org/wiki/False_discovery_rate#Benjamini.E2.80.93Hochberg_procedure| Benjamini and Hochberg].  For more information, see the  [http://www.broadinstitute.org/gsea/msigdb/help_annotations.jsp#overlap| Compute Overlaps Help] page.
+
     <li>Compute Overlap is now supports multiple hypothesis testing and outputs false discovery rate (FDR) q-values along the lines of  [http://en.wikipedia.org/wiki/False_discovery_rate#Benjamini.E2.80.93Hochberg_procedure| Benjamini and Hochberg] procedure, as well as hypergeometric p-values.  For more information, see the  [http://www.broadinstitute.org/gsea/msigdb/help_annotations.jsp#overlap| Compute Overlaps Help] page.
 
     </li>
 
     </li>
     <li>Modifying the Compute Overlaps procedure: Previously, gene universe (<strong>N</strong>) was defined as a union of the sets selected for testing and those from the input list.  This caused <strong>N</strong> to vary and also affected the final p-values.  The revised procedure defines the gene universe as all human gene symbols derived from the current version of the human Entrez Gene IDs, which provides a standard <strong>N = 46576</strong>, now used throughout Compute Overlaps.  For more information, see the [http://www.broadinstitute.org/gsea/msigdb/help_annotations.jsp#overlap| Compute Overlaps Help] page</li>
+
     <li>Compute Overlaps is now improved so that p-values are more readily comparable across collections. Previously, <strong>N</strong> (the population of genes, also known as the gene universe)) was defined as the genes in the sets selected for testing plus the genes in the input list.  This caused <strong>N</strong> to be different between runs of Compute Overlaps, so that the p-values were not always comparable between runs.  The revised procedure defines the gene universe as all human gene symbols derived from the current version of the human Entrez Gene IDs, which provides a standard <strong>N = 46576</strong>, now used throughout Compute Overlaps.  For more information, see the [http://www.broadinstitute.org/gsea/msigdb/help_annotations.jsp#overlap| Compute Overlaps Help] page</li>
     <li>Fixing broken links that caused a 404 error when trying to download a number of gene set collections from the [http://www.broadinstitute.org/gsea/msigdb/collections.jsp| About Collections] page.</li>
+
     <li>Fixing broken links for downloading some gene set collections from [http://www.broadinstitute.org/gsea/msigdb/collections.jsp| About Collections] page.</li>
 
     </li>
 
     </li>
 
</ul>
 
</ul>

Revision as of 16:40, 11 April 2013

<a href="http://www.broadinstitute.org/gsea/">GSEA Home</a> | <a href="http://www.broadinstitute.org/gsea/downloads.jsp">Downloads</a> | <a href="http://www.broadinstitute.org/gsea/msigdb/">Molecular Signatures Database</a> | <a href="http://www.broadinstitute.org/cancer/software/gsea/wiki/index.php/Main_Page">Documentation</a> | <a href="http://www.broadinstitute.org/gsea/contact.jsp">Contact</a>

Release 3.87 of the GSEA public Web site contains several updates and bug fixes, including:

  • Compute Overlap is now supports multiple hypothesis testing and outputs false discovery rate (FDR) q-values along the lines of Benjamini and Hochberg procedure, as well as hypergeometric p-values. For more information, see the Compute Overlaps Help page.
  • Compute Overlaps is now improved so that p-values are more readily comparable across collections. Previously, N (the population of genes, also known as the gene universe)) was defined as the genes in the sets selected for testing plus the genes in the input list. This caused N to be different between runs of Compute Overlaps, so that the p-values were not always comparable between runs. The revised procedure defines the gene universe as all human gene symbols derived from the current version of the human Entrez Gene IDs, which provides a standard N = 46576, now used throughout Compute Overlaps. For more information, see the Compute Overlaps Help page
  • Fixing broken links for downloading some gene set collections from About Collections page.