wiki:net.sf.basedb.reggie/notes423

Version 7 (modified by Nicklas Nordborg, 13 months ago) (diff)

Added note about sequence order for the second barcode read

Updating to Reggie 4.23


IMPORTANT!
This update requires a lot of manual work to install. Some things should be prepared in advance and some things should be done after Reggie 4.23 has been installed. Make sure to assign enough time for the update. No-one should be working with registrations and other things during the upgrade since some wizards may not work correctly until the update has been completed. It is also important to no secondary analysis jobs are running, since this may cause the results to be incorrectly registered or not registered at all.


1. New 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.

2. Changes in reggie-config.xml

There are several changes in reggie-config.xml:

  • The path to the pipeline scripts must be changed to point to the new version.
  • The path to Picard must be changed to a newer version: /usr/local/packages/picard-tools/2.20.0.
  • A new section for the MIPs demux has been added <demux-mips>. The default parameters can probably be used.
  • The parameter syntax for Picard has been changed. The old syntax key=value has been changed to -key value. This affects several existing configuration settings:
    • <demux>/<extract-options>
    • <demux>/<fastq-options>
    • <align>/<mark-duplicates-options>
    • <align-hisat>/<mark-duplicates-options>

3. Update BASE to BASE 3.15.2

This version is needed because it includes the Tag batch item importer that is needed in a later stop to import barcodes for the MIPs libraries.

4. Install Reggie 4.23 and run the installation wizard

5. Set a value for the Pipeline annotation on existing items

Almost all existing items from Library and downstream must be given a value for the Pipeline annotation. The table below specifies the value to set for the Pipeline annotation for different types of items. It should be possible to do a manual batch export/import for each row or groups of rows.

Item type Subtypes Additional filter Pipeline value
Extract Library, PooledLibrary, PooledLibraryAliquot - RNAseq
Physical bioassay FlowCell - RNAseq
Derived bioassay SequencingRun, DemuxedSequences, MergedSequences - RNAseq
Derived bioassay MaskedSequences, AlignedSequences Job type = HisatAlign RNAseq/Hisat/StringTie
Derived bioassay MaskedSequences, AlignedSequences Job type <> HisatAlign RNAseq/Legacy
Raw bioassay Cufflinks - RNAseq/Legacy
Raw bioassay StringTie - RNAseq/Hisat/StringTie
Protocol Demuxing, Merging - RNAseq
Software Demuxing, Merging - RNAseq

6. Other updates to existing items

Existing Hardware items of type=Sequencer should be updated with a value for the FlowCellType annotation. It should be either NextSeq or HiSeq.

7. New items that should be created

Item type Subtypes Additional settings Comment
Protocol DNANormalization TargetAmount, TargetMinimalAmount, TargetVolume, MIPS_Pool The protocol to use for normalizing DNA that should go into the MIPs pipeline. If no protocol exists, the option to pre-normalize DNA items (in the DNA/RNA/FlowThrough registration wizard) will not be enabled by default. Thus, it is recommended that a DNANormalization protocol is created only when needed.
Protocol Library preparation LibPrepTarget=MIPs One or more protocols that will appear in the "MIPs library registration" wizard
Protocol Pooling LibPrepTarget=MIPs One or more protocols that will appear in the MIPs pooling wizard
Protocol Demuxing Pipeline=MIPs One or more protocols that will appear in the MIPs demux wizard (demux section)
Software Demuxing Pipeline=MIPs One or more software entries that will appear in the MIPs demux wizard (demux section)
Protocol Merging Pipeline=MIPs One or more protocols that will appear in the MIPs demux wizard (merge section)
Software Merging Pipeline=MIPs One or more software entries that will appear in the MIPs demux wizard (merge section)

8. Barcodes and barcode template plates

Most of the barcode information (such as the sequences, positions, etc.) should be available as a file that can be used with batch importers. Some steps need to be done manually.

  1. Import/create Tag items from file using the batch Tag importer and then add annotations with the Annotation importer. The following values are required:
Name Name of the barcode/tag which must be unique
Pipeline All items should be set to MIPs
BarcodeSequence The barcode sequence for the first index read
BarcodeSequence2 The barcode sequence for the second index read

NOTE! The sequences for the second barcode are expected to be in the order that corresponds to the NextSeq sequencer. If the sequencing has been done on a HiSeq they sequences may need to be reverse-complemented before used in the demux. See https://support.illumina.com/content/dam/illumina-support/documents/documentation/system_documentation/miseq/indexed-sequencing-overview-guide-15057455-05.pdf for more information.

  1. A barcode template is represented as a Bioplate item where each coordinate is linked to a Tag item. Start by manually creating the required number of bioplates. They should have the following values:
Bioplate type BarcodeTemplate
Plate geometry 96-well (8x12)
Pipeline (annotation) MIPs
  1. Link the barcodes to templates. This is done by creating (virtual) extracts and link them with tags and positions on one of the bioplates created in the first step. It should be possible to do this with the batch Extract importer:
Name Name of the extract (recommended to be the same as the name of the Tag)
Tag Name of the tag
Bioplate Name of the template bioplate
Biowell row Row coordinate on the template bioplate (A-H)
Biowell column Column coordinate on the template bioplate (1-12)

9. User accounts for MIPs pipeline

A user account that should work with the MIPs pipeline need to have the following group/role memberships:

  • Groups: SCAN-B Lab, MIPsSecondaryAnalysis
  • Roles: User, MIPsLibraryPrep, MIPsPlateDesigner