= Updating to Reggie 4.32 =

**Note that BASE 3.19 is required for running Reggie 4.32.** 

== A. Before updating Reggie ==

=== 1. Updated pipeline script ===

The `singlecolumnaverager.awk` script has been updated (see #1325). Install the new version at a suitable location and make sure to update reggie-config.xml with the new path.

=== 2. Install VCF files for targeted genotyping ===

VCF files (`dpyd.vcf`, `esr1.vcf` and `pik3ca.vcf`) for the targeted genotyping are in the http://onk-wiki.bmc.lu.se/svn/scanbprim/support-files/variant-calling/targeted repository and should be installed in the same location as the other databases used in the variant calling process. The VCF files need to be indexed by running the `index_all.sh` script. Finally, the `targeted-genotyping.toml` is a also a new file.


== B. Update Reggie ==

Update Reggie to version 4.32 as usual, but note that several additions are needed in `reggie-config.xml` so a bit of preparation is maybe needed. Do not forget to run the Installation wizard. This will create and fix several new annotations, etc. It will still complain about File server items that need to be manually configured:

  * !ImportGateway: This should be configured to access `lida1:/casa17/cmdtransfer` as the `scanb` user.
  * !ImportArchive: This should be configure to access `lida1:/casa17/project_archive/import_archive/` as the `scanb` user.

The same paths also need to be specified in the `reggie-config.xml` file in `<import-gateway>` and `<import-archive>` settings.

Also, a new entry `<step-1-import>` is needed since the first Trimmomatic step is different in the import pipeline.

There are also several new configuration entries related to the targeted genotyping. See #1323.

 * A new section `<targeted-genotyping>` defining the 3 currently existing targets: DPYD, ESR1 and PIK3CA
 * A new entry in the `<programs>` section is needed for `<gatk4>`.
 * Several new entries in the `<variant-call>` section.
   * `<vcfanno-options-targeted>` 
   * `<snpeff-options-targeted>`
   * `<haplotypecaller-options>`

== C. After updating Reggie ==

=== 1. Change existing software items ===

 * All `Demuxing` software should be annotated with `DemuxType=Picard`
 * All `Merging` software should be annoteted with `MergeType=Default`

=== 2. Create new software items ===

 * Create a new `Demuxing` software representing the bcl2fastq demuxing that is done by CMD. Annotate `DemuxType=bcl2fastq`, `Pipeline=RNAseq`, `ExternalRef=bcl2fastq` and make it project default.

 * Create a new `Merging` software representing the FASTQ import variant that uses a different Trimmomatic first step. Annotate `MergeType=Import`, `Pipeline=RNAseq` and make it project default.

 * Create a new `Variant calling` software item. Annotate with `VariantCallType=TargetedGenotype` and make it the project default.


=== 3. Create or annotate protocol items ===

We either need to create new protocol items or annotate existing protocol items with `ExternalRef` so that they can be referenced in imports from CMD. 

 * Library preparation
 * Allprep extraction (3 variants for RNA, DNA and !FlowThrough)
 * Sample handling


=== 4. Fix pool aliquot names ===

Pool aliquots created by the external sequencing wizard have incorrect names. See #1301.

=== 5. Run the targeted genotyping analysis for all existing variant calls ===

The **Start targeted genotyping** wizard is in the **RNA library preparation and analysis** section inside the **Secondary analysis wizards** box.