report erros

[Abhinav Seth]

Hi,

I've run several successful jobs today.

However some of the jobs this evening have been running but failing to produce graphical output. I get the error message listed below.

Do you have any ideas on what might be causing this?

Thanks,

Abhi

The Tool ran successfully but at least one part of the reports production failed with the following details The reports is INcompleteERROR 1 Must be finite. ---------------------------------------------------java.lang.IllegalArgumentException: Must be finite. at jfree...@1.5.0/org.jfree.chart.axis.NumberTickUnitSource.getCeilingTickUnit(NumberTickUnitSource.java:114) at jfree...@1.5.0/org.jfree.chart.axis.NumberAxis.selectHorizontalAutoTickUnit(NumberAxis.java:852) at jfree...@1.5.0/org.jfree.chart.axis.NumberAxis.selectAutoTickUnit(NumberAxis.java:825) at jfree...@1.5.0/org.jfree.chart.axis.NumberAxis.refreshTicksHorizontal(NumberAxis.java:954) at jfree...@1.5.0/org.jfree.chart.axis.NumberAxis.refreshTicks(NumberAxis.java:926) at jfree...@1.5.0/org.jfree.chart.axis.ValueAxis.reserveSpace(ValueAxis.java:806) at jfree...@1.5.0/org.jfree.chart.plot.XYPlot.calculateDomainAxisSpace(XYPlot.java:3031) at jfree...@1.5.0/org.jfree.chart.plot.XYPlot.calculateAxisSpace(XYPlot.java:2990) at jfree...@1.5.0/org.jfree.chart.plot.XYPlot.draw(XYPlot.java:3132) at jfree...@1.5.0/org.jfree.chart.JFreeChart.draw(JFreeChart.java:1286) at jfree...@1.5.0/org.jfree.chart.JFreeChart.createBufferedImage(JFreeChart.java:1453) at jfree...@1.5.0/org.jfree.chart.JFreeChart.createBufferedImage(JFreeChart.java:1434) at jfree...@1.5.0/org.jfree.chart.ChartUtils.writeChartAsPNG(ChartUtils.java:184) at jfree...@1.5.0/org.jfree.chart.ChartUtils.saveChartAsPNG(ChartUtils.java:310) at jfree...@1.5.0/org.jfree.chart.ChartUtils.saveChartAsPNG(ChartUtils.java:285) at org.gsea_msigdb.gsea/edu.mit.broad.genome.charts.XChartImpl.saveAsPNG(XChartImpl.java:114) at org.gsea_msigdb.gsea/edu.mit.broad.genome.reports.EnrichmentReports.createGseaLikeReport(EnrichmentReports.java:317) at org.gsea_msigdb.gsea/edu.mit.broad.genome.reports.EnrichmentReports.createGseaLikeReport(EnrichmentReports.java:179) at org.gsea_msigdb.gsea/xtools.gsea.AbstractGsea2Tool.execute_one_with_reporting(AbstractGsea2Tool.java:124) at org.gsea_msigdb.gsea/xtools.gsea.Gsea.execute(Gsea.java:165) at org.gsea_msigdb.gsea/xtools.api.AbstractTool.module_main(AbstractTool.java:417) at org.gsea_msigdb.gsea/org.genepattern.modules.GseaWrapper.main(GseaWrapper.java:287) at org.gsea_msigdb.gsea/xapps.gsea.CLI.main(CLI.java:29)

[Anthony Castanza]

Hi Abhi,

 

The crux of the error is here: java.lang.IllegalArgumentException: Must be finite

Then if you look at the error message you’ll see where exactly it failed “org.jfree.chart.axis.NumberAxis.selectHorizontalAutoTickUnit”

 

What this is saying is that when its trying to generate the axis dimensions to plot your gene rankings its encountering an infinite value that it can’t plot as a real finite value on the axis.

I assume this was in a GSEA job? GSEA Preranekd? If so, you’ll need to check your ranked list and resolve any non-numeric values. Some people replace infinite values with a small number above/below the positive/negative max value.

 

-Anthony

 

Anthony S. Castanza, PhD

Curator, Molecular Signatures Database

Mesirov Lab, Department of Medicine

University of California, San Diego

http://gsea-msigdb.org/

--
You received this message because you are subscribed to the Google Groups "GenePattern Help Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genepattern-he...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/genepattern-help/cfb7d79a-6848-424f-9556-7058cdce8dean%40googlegroups.com.

[Abhinav Seth]

Hi,

That is very helpful. This is a GSEA job, not preranked. When I open the ranked list, I do see some genes have infinity score. When I look back at my dataset, these are genes where one of the categories had zero expression across all samples. What is the recommended way to deal with this if I want to complete the analysis as a GSEA job and not use the ranklist to run GSEA preranked?

Thanks

[Anthony Castanza]

How many samples per group do you have, and what metric for ranking genes are you using?

If you have three or more samples per group you should be able to run using the signal2noise ranking metric which has a built in mitigation this issue.

If you're using some other ranking metric, and aren't able to use signal2noise, I'd suggest adding a small pseudo-count to your data.

-Anthony

Anthony S. Castanza, PhD
Curator, Molecular Signatures Database
Mesirov Lab, Department of Medicine
University of California, San Diego

--

You received this message because you are subscribed to the Google Groups "GenePattern Help Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genepattern-he...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/genepattern-help/baa700e6-cb6b-4a8d-9f5f-cc422a57f77an%40googlegroups.com.

[Abhinav Seth]

Unfortunately it is somebody else's published data and there are only two replicates per condition, so I am unable to use signal2noise. I am trying to run GSEAprerank now, if that doesn't work, i'll try the pseudo count approach. It's weird that it's only doing it for some of the genes with zero values.

Thanks for your help.