Gsea enhancements

From GeneSetEnrichmentAnalysisWiki

(Difference between revisions)
Jump to: navigation, search
 
(16 intermediate revisions not shown)
Line 1: Line 1:
-
<span style="font-weight: bold;">Note</span>: By default the hyperlinks from the software will point to the PROD server. To make them connect to the DEV server, use the -Ddebug=true flag <span style="color: rgb(255, 0, 255);">[Keep on enh/bug page.]</span><br /><br /><span style="font-weight: bold;">Beta testing 4/18:</span><br /><br />1. Created a tiny dataset with 4 samples; create phenotype on the fly with 1 sample in ClassA and 3 in ClassB; got this error. If I create phenotype on the fly with 2 samples in each class, life is good.  <error details="">  ---- Full Error Message ---- Nan hit score for feature: TACC2  ---- Stack Trace ---- # of exceptions: 1 ------Nan hit score for feature: TACC2------ java.lang.IllegalStateException: Nan hit score for feature: TACC2 at edu.mit.broad.genome.alg.gsea.KSCore.calculateKSScore_all_modes(KSCore.java:137) at edu.mit.broad.genome.alg.gsea.KSCore.calculateKSScore(KSCore.java:46) at edu.mit.broad.genome.alg.gsea.KSTests.shuffleTemplate_canned_templates(KSTests.java:377) at edu.mit.broad.genome.alg.gsea.KSTests.shuffleTemplate(KSTests.java:292) at edu.mit.broad.genome.alg.gsea.KSTests.executeGsea(KSTests.java:156) at edu.mit.broad.genome.alg.gsea.KSTests.executeGsea(KSTests.java:130) at xtools.gsea.Gsea.execute_one(Gsea.java:152) at xtools.gsea.Gsea.execute(Gsea.java:104) at edu.mit.broad.xbench.tui.TaskManager$ToolRunnable.run(TaskManager.java:464) at java.lang.Thread.run(Unknown Source) 2. From the command line, gene set names have to be case sensitive. They should be case INsensitive. (Tested using xtools.gsea.LeadingEdgeTool, where -gsets is comma-separated list of gene set names.)<span style="font-weight: bold;"><br /><br /></span><span style="color: rgb(255, 102, 0);">If 2 samples or less in a class then signal to noise, tTest will not work. Need to use ratio of means.</span><span style="font-weight: bold;"><br /></span><span style="color: rgb(255, 0, 255);">If you want to trap the error and provide a &quot;standard&quot; error box with a help button; I'm happy to write wiki text for it. Let me know.</span><span style="font-weight: bold;"><br /><br /></span></error><error details=""><span style="font-weight: bold;">Found 4/4:</span><br />1. Specify Bhattacharyya with Continous pheno, should get error 1011 (get hardcoded error). <span style="font-style: italic;">Metric for ranking genes</span> parameter.<br /><br /><span style="color: rgb(0, 0, 255);">Bhattacharyya is a continuous metric so isnt this correct?<br /><span style="color: rgb(255, 0, 0);">When I run </span></span><span style="color: rgb(255, 0, 0);">Bhattacharyya with a continuous phenotype I get the following error (no error help button):<br /><br /></span><span style="color: rgb(0, 0, 255);"><span style="color: rgb(255, 0, 0);">&nbsp;Tool execution error</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">&nbsp;Message: Template is not biphasic. Name: 100_g_at_profile_in_p53_dataset_hgu95av2.cls#100_g_at # splits= 50</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">&nbsp;This metric can only be used with 2 class comparisons</span><br style="color: rgb(255, 0, 0);" /><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">-------------------------------------------------------------</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">java.lang.RuntimeException: Template is not biphasic. Name: 100_g_at_profile_in_p53_dataset_hgu95av2.cls#100_g_at # splits= 50</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">&nbsp;This metric can only be used with 2 class comparisons</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">&nbsp;&nbsp;&nbsp; at edu.mit.broad.genome.alg.VectorSplitter._barf_not_biphasic(VectorSplitter.java:71)....</span><br /></span><br />2. Specify Pearson with Categorical, should get error 1010 (get hardcoded error). <span style="font-style: italic;">Metric for ranking genes</span> parameter.<br /><br /><span style="color: rgb(0, 0, 255);">Pearson is allowed for categorical &amp; continuous<br /><span style="color: rgb(255, 0, 0);">When I run Pearson with a Categorical phenotype, I get the following error (no help button); now that I look more closely, it doesn't seem to be related to the Categorical/Continuous thing...<br /><br style="color: rgb(255, 0, 0);" /></span><span style="color: rgb(255, 0, 0);">Message: For input string: &quot;MUT&quot;</span><br style="color: rgb(255, 0, 0);" /><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">-------------------------------------------------------------</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">java.lang.NumberFormatException: For input string: &quot;MUT&quot;</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">&nbsp;&nbsp;&nbsp; at sun.misc.FloatingDecimal.readJavaFormatString(Unknown Source)</span><br style="color: rgb(255, 0, 0);" /><span style="color: rgb(255, 0, 0);">&nbsp;&nbsp;&nbsp; at java.lang.Float.parseFloat(Unknown Source)....</span><br /></span></error><error details=""><span style="font-weight: bold;"><br />Found/requested 3/22 (with installed Gsea2):<br /></span><br /><span style="color: rgb(51, 153, 102);"> Run GSEA, add a phenotypes file, create phenotypes on the fly (works fine), click the Show Phenotypes from all Sources. Should show labels from both the file you added and the one you just created, but now shows only one at a time. (This seems to have broken in the 3/21 build; I'm pretty sure it was working in the 3/20 build).<br /><span style="color: rgb(255, 0, 255);">Open&nbsp; (4/24) </span><br /></span></error><span style="color: rgb(255, 102, 0);"></span>
+
<p><span style="font-weight: bold;"><br /></span>We are now using [http://bugzilla.broad.mit.edu/gsea/ Bugzilla] to record and track GSEA bugs. <br /></p>
 +
<p>Bugs from this page that were verified/rejected before we started using Bugzilla, can be found in  [http://wwwdev.broad.mit.edu/gsea/doc/GSEA_enh_history.doc GSEA_enh_history.doc].<br /><span style="font-weight: bold;"></span><span style="color: rgb(255, 0, 255);"></span><br /></p>

Current revision as of 19:04, 16 May 2006


We are now using Bugzilla to record and track GSEA bugs.

Bugs from this page that were verified/rejected before we started using Bugzilla, can be found in GSEA_enh_history.doc.

Personal tools