wiki:net.sf.basedb.reggie/notes425

Version 2 (modified by Nicklas Nordborg, 8 months ago) (diff)

Update the release notes for Reggie 4.25

Updating to Reggie 4.25

A. Before updating Reggie

1. Disable the Variant calling pipeline

Disable auto-confirmation for the variant calling by setting AutoProcessing=Disable on the Variant Calling Pipeline item list.

2. Install new/updated pipeline scripts

There are new pipeline scripts: http://baseplugins.thep.lu.se/browser/other/pipeline/trunk Download and install at a suitable location on the prime cluster:

  • mut_stats.py
  • mutation_signature.R
  • Rscript.sh

3. Install new databases for annotating variants

The following databases have been updated:

  • dbSNP to version 153
  • COSMIC to version 90
  • gnomAD to version 2.1.1
  • Swegen to version 20180409

The other databases are the same as before. The database update also affect some of the other configuration files. Everything that is needed can be found at casa4:/disk/lv-p822a-50tb-b/thep-nni/vcall/newdbs. All files and subdirectories should be copied to a new location in the "references" directory in the Prime cluster, for example, prime:/disk/reference/scanb/rnaseqvarcall-feb2020.

Change the configuration <variant-call><base-dir> setting in reggie-config.xml to point to the new location of the databases.

B. Update Reggie

Update Reggie to version 4.25 as usual. Do not forget to run the Installation wizard.

C. After updating Reggie

1. Create new SCAN-B databases

Use the new Reggie wizard (Build SCAN-B variant database) to create new databases for the existing SCAN-B samples. The wizard need to create two databases, one with tumor samples and one with normal samples. This may take a few hours. Databases will be saved in the temporary job directory (the exact path is printed in the job information afterwards) and need to moved to the database-directory manually, eg. prime:/disk/reference/scanb/rnaseqvarcall-feb2020/scanb

Note that both databases should be stored in the same directory, and make sure that they have correct names:

  • scanb-tumors.vcf.gz (and a corresponding .tbi file)
  • scanb-normals.vcf.gz (and a corresponding .tbi file)

2. Test the variant calling pipeline

The annotation and filtering part of the variant calling pipeline can now be tested. Select some alignments that already have raw variant calls.

3. Enable the Variant calling pipeline

Enable auto-confirmation for the variant calling by removing AutoProcessing=Disable from the Variant Calling Pipeline item list.

4. Complete the annotation and filtering step for all alignments that already have raw variant calling