comparison QDNAseq.xml @ 27:f89205f51e27 draft default tip

Uploaded

26:069289631381

  <command interpreter="Rscript"> 
    QDNAseq.R 
    $qdnaseq_cfg
    \$QDNASEQ_PATH
  </command>

  <stdio>
    <!-- Anything higher than 0 means the R script didnt finish (correctly) -->
    <!-- Because different R packages deal with err/warn differently unable to waterproof this -->
    <exit_code range="1:" level="fatal" description="R script finished too early, check log" />

    <data format="html" name="htmlFile" label="QDNAseq: ${jobName} report" />

    <!-- WHY does there seem to be no way to use split() within this code in galaxy!!! -->
    <!-- now have to fall back to using unique names within binSizes instead of just integers -->
    <!-- Problem with integers is that both "1" and "5" are also present in eg "15,100" -->
    <data format="tsv" name="txt_1000" label="QDNAseq: ${jobName} txt 1000kb">
      <filter>( "bin1000kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tsv" name="txt_100" label="QDNAseq: ${jobName} txt 100kb">
      <filter>("bin100kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tsv" name="txt_30" label="QDNAseq: ${jobName} txt 30kb">
      <filter>("bin30kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tsv" name="txt_15" label="QDNAseq: ${jobName} txt 15kb">
      <filter>("bin15kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tsv" name="txt_5" label="QDNAseq: ${jobName} txt 5kb">
      <filter>("bin5kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tsv" name="txt_1" label="QDNAseq: ${jobName} txt 1kb">
      <filter>("bin1kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>

    <data format="rds" name="rds_1000" label="QDNAseq: ${jobName} rds 1000kb">
      <filter>( "bin1000kb" in binSizes and rds2history == 'TRUE')</filter>

The input BAMs are expected to be **single end reads of 50bp length** mapped to GRCh37/hg19 genome build. Other experiment setups are currently not tested or supported within galaxy. See the documentation of QDNAseq at bioconductor on how to deal with different setups (or keep fingers crossed ;) )

.. class:: warningmark

Requires **internet access** for downloading bin-annotations from bitbucket and to show some styling (css) of the final report

-----

**Citation**

27:f89205f51e27

  <command interpreter="Rscript"> 
    QDNAseq.R 
    $qdnaseq_cfg
    \$QDNASEQ_PATH
  </command>

  <version_command interpreter="Rscript">QDNAseq-version.R</version_command>

  <stdio>
    <!-- Anything higher than 0 means the R script didnt finish (correctly) -->
    <!-- Because different R packages deal with err/warn differently unable to waterproof this -->
    <exit_code range="1:" level="fatal" description="R script finished too early, check log" />

    <data format="html" name="htmlFile" label="QDNAseq: ${jobName} report" />

    <!-- WHY does there seem to be no way to use split() within this code in galaxy!!! -->
    <!-- now have to fall back to using unique names within binSizes instead of just integers -->
    <!-- Problem with integers is that both "1" and "5" are also present in eg "15,100" -->
    <data format="tabular" name="txt_1000" label="QDNAseq: ${jobName} txt 1000kb">
      <filter>( "bin1000kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tabular" name="txt_100" label="QDNAseq: ${jobName} txt 100kb">
      <filter>("bin100kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tabular" name="txt_30" label="QDNAseq: ${jobName} txt 30kb">
      <filter>("bin30kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tabular" name="txt_15" label="QDNAseq: ${jobName} txt 15kb">
      <filter>("bin15kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tabular" name="txt_5" label="QDNAseq: ${jobName} txt 5kb">
      <filter>("bin5kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>
    <data format="tabular" name="txt_1" label="QDNAseq: ${jobName} txt 1kb">
      <filter>("bin1kb" in binSizes and txt2history == 'TRUE')</filter>
    </data>

    <data format="rds" name="rds_1000" label="QDNAseq: ${jobName} rds 1000kb">
      <filter>( "bin1000kb" in binSizes and rds2history == 'TRUE')</filter>

The input BAMs are expected to be **single end reads of 50bp length** mapped to GRCh37/hg19 genome build. Other experiment setups are currently not tested or supported within galaxy. See the documentation of QDNAseq at bioconductor on how to deal with different setups (or keep fingers crossed ;) )

.. class:: warningmark

Some optional history input/output files are of format "rds" (file format to store a R object). This is not registered in galaxy by default, so has to be added to the available datatypes.

-----

**Citation**