Skip to content

4.2.1
Compare
Choose a tag to compare
@RoanKanninga RoanKanninga released this 31 Aug 07:10
· 63 commits to master since this release
4.2.1
f340d04
This commit was created on github.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

Dragen

  • added new protocols
    • PreparedDragenData
    • CombineDragenSampleData
    • ReheaderVcf
  • new workflow_DRAGEN
  • new generate_DragenScripts

Bucketsystem

From this version on the folders Samplesheets, tmp, project and generatedscripts have a subfolder with the name of the pipeline (e.g. NGS_DNA, NGS_Demultiplexing,VIP, GAP etc). This has been implemented due to the fact that one machine does both the rawdata handling (AGCT/NGS_Demultiplexing) and the variantcalling (GAP/NGS_DNA). Data archiving is still on a seperate machine (chaperone)

Getting rawdata from permanent storage changed

Diagnostic clusters cannot talk directly to permanent storage anymore (i.e. permanent storage can push only).

  • Created new protocol: CheckRawDataOnTmp
    this protocol comes instead of CopyPrmToTmpData, it will check whether the rawdata is already available, if not than a file will be created in the logs folder as ${logsDir}/${project}/${project}.data.requested, this filetype will be checked by a new NGS_Automated protocol (copyRawDataToTmp) and will push the data to the diagnostic cluster

coverage calculations per gVCF (instead of per BAM)

This script will use gVCF files as input, due to all kind of possible rearrangements of the reads during the variantcalling the bam files created by the pipeline are not always the end product. Therefore it is more accurate to calculate coverage based on the variantcaller created gVCF files.
End product is not changed, still in this format:
"Index\tChr\tChr Position Start\tChr Position End\tAverage Counts\tDescription\tReference Length\tCDS\tContig"
But additional file(s) are created to calculate percentages/bases with zero coverage, dp<20, GQ values (.{bedfile}.CoverageOutput.csv)

Diploid chrX

  • Male chromosome X will be called diploid
    This means that there is no need to have 2 seperate chr X batches (nonPar and Par))

updated tool versions:

bcfTools 1.16
BEDTools 2.30.0
CADD-v1.4
GATK 4.2.4.1
gVCF2BED 1.0.0 (new tool)
HTSlib 1.16
jq/1.6
multiQC-1.12
ngs-utils 23.04.1
picard 2.26.0
Python 3.10.4
R 4.2.1
SAMTools 1.16.1
vcfAnno v0.3.3

Others

  • moved unused protocols to deprecated and removed them from the workflow.csv
    vcfToTable
    CoverageCalculations
    XHMM
    Convading
    DecisionTree
Assets 2
Loading