Skip to content

Latest commit

 

History

History
67 lines (47 loc) · 4.36 KB

settings.md

File metadata and controls

67 lines (47 loc) · 4.36 KB

Settings

Section 'tools'

Paths of locally installed tools.
Paths are preset and should not be changed unless you want to exchange a tool.

Section 'tools-data'

Paths of downloaded data used by specific tools. Paths are preset and should not be changed unless you want to change where the data is stored.

  • ngs-bits_local - the path to a local ngs-bits installation can be provided here. If no path is provided the ngs-bits apptainer container listed in the 'apptainer-container' section is used.

Section 'apptainer-container'

Paths of tools encapsulated in apptainer containers.
Paths are preset and should not be changed unless you want to exchange a container.

Section 'folders'

This section contains path to important folders:

  • data_folder - general data folder with downloaded database flat files.
  • local_data - local data folder. This folder is used to copy database files for annotation from data_folder to a local temporary folder. The local temporary folder is usually on a SSD/NVME, so this will speed up the annotation.
  • container_folder - general container folder with downloaded apptainer container images. Images can be copied to a subdirectory of locla_data to improve access speed.
  • test_data_folder - folder for test data. This folder is needed for unit tests and tool tests during pipeline development.
  • project_folder - folder that contains the NGS data. This folder has to be set for the NGSD integration (see below). Instead of providing a single path, it is also possible to provide project-type-specific paths - in this case the dictionary syntax of PHP is used, e.g. project_folder['diagnostic'] = /mnt/projects/diagnostic/

Section 'general'

This section contains general settings:

  • copy_dbs_to_local_data - Flag (true/false) that indicates if large database files (gnomAD, etc) and apptainer container are copied to the local_data folder.
  • delete_fastq_files - Flag (true/false) that indicates if FASTQ files are to be deleted after the BAM/CRAM file is created after mapping.
  • cnv_bin_size_wgs - Bin size used for CNV analysis of WGS samples.
  • cnv_bin_size_shallow_wgs - Bin size used for CNV analysis of shallow WGS samples.
  • cnv_bin_size_longread_wgs - Bin size used for CNV analysis of long-read WGS samples.
  • use_bwa1 - Flag (true/false) that indicates if BWA mem should be used for mapping instead of BWA mem 2 (BWA mem 2 is faster, but needs more RAM).
  • bwa_mem2_suffix - Suffix appended to the BWA-mem2 executable, e.g. .avx2 for AMD EPYC9654.
  • annotate_refseq_consequences - Flag (true/false) that indicates if variant consequences based on RefSeq transcripts should be annotated in addition to variant consequences based on Ensembl transcripts.
  • custom_columns - Used to add custom annotations to the output VCF/GSvar file. Each entry consists of a colon-speparated list of VCF file, INFO field name in the source VCF (prefixed with CUSTOM_ in the annotated VCF) and column description. Provide the annotation using the dictionary syntax of PHP, e.g. custom_columns['NGSD_counts'] = "/mnt/data/dbs/NGSD/NGSD_germline.vcf.gz;COUNTS;NGSD counts".
  • locaton - If set enables site-specific functionality and tests. Set only if you are a collaborator with a specific site name.

Section 'dragen'

This section contains settings to run the germline/somatic data analysis on a on-site Illumina Dragen server:

  • dragen_in - Folder into which FASTQ data is copied for Dragen data analysis.
  • dragen_out - Folder into which Dragen output files are copied after the analysis.
  • dragen_data - Folder used as working directory on the Dragen server.
  • dragen_genomes - Folder containing genome files for Dragen on the Dragen server.
  • dragen_log - Folder used to write SGE logs of Dragen analyses.
  • queues_dragen - Comma-separated list of SGE queues for Dragen servers (one per server).
  • use_dragen_sv_calling - Flag (true/false) that indicates if SV calls produced by Dragen are used. If set to false, SV calling based on Manta is used.

Section 'mysql-databases'

This section contains NGSD data (host, database name, credentials).

Section 'grid_engine'

This section contains settings for automated data analysis via SGE:

  • queues_default - Default SGE queues.
  • queues_research - Research SGE queues.
  • queues_high_priority - High-priority queues.
  • queues_high_mem - High-memory consumption queues (RNA).