Skip to content

README.md

Latest commit

 

History

History
83 lines (51 loc) · 3.47 KB

README.md

File metadata and controls

83 lines (51 loc) · 3.47 KB

TSO500_post_processing

Documentation

The Illumina TSO500 manual can be found on Q-Pulse

EI-GEN-LocalAppUserGuide_TSO500

To run the pipeline

To start the pipeline from demultiplexing, 1_TSO500.sh script should be copied into the run folder (/data/output/results/runid/TSO500/). From this folder run the command:

sbatch --export=raw_data=/data/raw/novaseq/<run_id>/ 1_TSO500.sh

The raw data directory must contain the SampleSheet.csv.

To re-run DNA samples with a new referral

The script rerun_coverage.sh allows the coverage json file to be produced for a DNA sample with a new referral in all scenarios (analysis/ folder present or deleted). This can be run as follows, as transfer, from any location:

sbatch rerun_coverage.sh <run_id> <sample_id> <worksheet> <new referral>

The new coverage file will then be present in Gathered_Results/Database/ alongside a new samples list file for upload to the database.

Duty scientist responsibilites

The duty scientist is responsible for the following tasks:

  • Creating the TSO500 samplesheet- see below for requirements
  • Signing off the run in autoqc database - for more information refer to the AutoQC sop
  • Importing the data into the data into the somatic variant database

Samplesheet requirements

  • The samplesheet must contain the samples in the correct order for the RNA contamination results to be valid
  • Every NTC must be named NTC-worksheetid

Unit tests

Unit tests have been created against the following scripts: tsv2db.py, coverage2json.py, fusion_check_with_ntc.py.

To run all unit tests:

  • copy these scripts into the tests/ folder (relative imports not currently working, will be fixed in future version)
  • activate the TSO500_post_processing conda environment
  • change into tests directory: cd tests/
  • run python -m unittest

To run tests on a specific script, follow the steps above but run python -m unittest <test_script_name>

Adding a new DNA panel

To add a new panel, the following needs to be changed:

Samplesheet generator:

  • Add the referral reason to the samplesheet generator (see SOP in Qpulse) and make sure it matches the filename of the new bed files (case sensitive)

Pipeline:

  • Generate bed files for the new panel
    • The hotspot_variants/*bed bed file and hotspot_coverage/*combined.bed files are required
    • The hotspot_coverage/*hotspots.bed and hotspot_coverage/*genescreen.bed files are optional
    • Filenames should be all lowercase
  • Make sure that all regions in the new panel are covered in vendorCaptureBed_100pad_updated.bed (the bed file file used to generate the depth of coverage file)
  • Make sure that any flanking regions are added to the TSO_extra_padding_chr.interval_list file - Illumina bed file only goes +/- 2bp so this file contains the extra 3bp to make it +/- 5bp

Somatic variant database:

  • Make a new panel object in the somatic variant database that matches the filename of the new bed files (case sensitive)
  • Move the new variants bed file into the roi/variant_calling folder in the somatic database

Adding a new RNA panel

To add a new panel, the following needs to be changed:

Samplesheet generator:

  • Add the referral reason to the samplesheet generator (see SOP in Qpulse)

Pipeline:

  • Add a new file to RNA_referrals named .txt with the gene names on the panel, one per line

Somatic variant database:

  • Make a new panel object in the somatic variant database that includes the genes on the panel to filter fusions by