Home/
Set Analysis Parameters

Set Analysis Parameters

1. Open DRAGEN Germline Pipeline from BaseSpace™ Sequence Hub as follows.
  1. Select the Apps tab, and then select DRAGEN Germline Pipeline.
  2. From the Version drop-down list, select 3.3.7.
  3. Select Launch Application.
2. To override the default analysis name, enter a preferred analysis name in the Analysis Name field.

The default is the app name with the date and time the session started.

3. From the Save Results To field, select Select Project, and then select a project to store app results to.
4. Specify a sample by selecting the option that matches the input file type. Multiple samples can be selected in a single row.
5. [Optional] From the Sample Sex drop-down list, select the sex of the sample.

If you plan on using Expansion Hunter for the analysis, you must specify the sex.

6. Set the analysis pipeline configuration.
Map/Align Only–Samples are mapped and aligned to the reference genome and position-sorted.
Map/Align + Small Variant Caller–In addition to the Map/Align Only processes, variant calling is performed.
Small Variant Caller–Only variant calling is performed. This configuration only accepts BAM inputs files.
7. From the Reference drop-down list, select a reference genome.
8. If you selected Custom from the Reference drop-down list, select the custom DRAGEN and/or FASTA reference files.
DRAGEN–The custom reference file must have been generated by the DRAGEN Reference Builder v3.x apps.
FASTA–Used for Expansion Hunter or SV.
9. From the Map/Align Output drop-down list, specify whether to output a BAM, CRAM, or no alignment file at all.
10. From the Variant Caller Output drop-down list, specify the VCF output type:
VCF
GVCF–Variants are recorded individually and nonvariants are grouped into blocks.
GVCF with BP_RESOLUTION–Variants and nonvariants are recorded individually. This option will increase run time and create large gVCF files, eg, 2 hours for a 30x sample with a 20 GB gVCF.
11. [Optional] If you enabled a variant caller for the analysis, select a target BED file.

The contig names must match those of the chosen reference. If a mismatch is detected, variant calling will be abandoned.

12. Specify regions over which to produce coverage metrics as follows:
  1. Select Select Dataset File(s).
  2. Select the BED file(s) that contain the regions for which you want to produce coverage metrics.
  3. Enter a MAPQ filter value. Any read with a MAPQ value less than this threshold will be filtered out.
  4. Enter a BQ filter value. Any base call with a quality score less than this threshold will be filtered out.
  5. Set the Full-Res option. Disabled by default, enabling this option will generate large files.
13. Specify additional run settings by expanding the function headings and selecting the appropriate checkboxes.

Heading

Description

CNV

Enables germline CNV calling. If enabled, select the Segmentation Algorithm.

Circular Binary Segmentation–Iteratively identifies change points in a genomic sequence using a nonparametric hypothesis testing approach.
Shifting Level Model–Models genomic data as the sum of two independent stochastic processes and segments using a subclass of Hidden Markov Model.

SV (Manta)

Enables SV (Manta) analysis.

Expansion Hunter

Enables calling of repeat-expansion variants.

Advanced Settings

Duplicate Marking–Enabled by default.
BQD–Enable base quality drop off detection. Enabled by default.
Depth of Coverage–Enables depth-of-coverage calculations. A target BED file is required, and coverage calculations are performed over the regions definited in the BED file. *genomecov.bed file is generated.
dbSNP VCF–Select the variant annotation database .vcf or .vcf.gz file.

Automation Settings

Enables automation settings. Specify a sample by selecting the option that matches the input file type and the sex.

14. Select Launch Application to start the analysis.

When the analysis is complete, the status of the app session is automatically updated and you receive a confirmation email.