Workbench & Pipeline

1. AI shortlist (38.0+)

The AI Shortlist Module, located in the Workbench & Pipeline section of Organization Settings, allows you to refine variant prioritization based on analysis type.

1.1 Rare Diseases

The AI shortlist prioritizes genes of known and unknown significance within the shortlist.

• Focused Mode: The AI shortlist will prioritize known genes and unknown genes separately, and will be limited to likely solving variants.

• Discovery Mode: The AI shortlist will prioritize known and unknown genes, and will be limited to likely solving variants.

1.2 Carrier

The AI shortlist will prioritize variants reported as pathogenic or likely pathogenic in any known variant databases or variants with high severity.

• Known Pathogenic: The AI shortlist will prioritize variants reported as pathogenic or likely pathogenic in any known variant databases.

• High Severity: The AI shortlist will prioritize variants with high severity.

• Both: The AI shortlist will prioritize variants reported as pathogenic or likely pathogenic in any known variant databases or variants with high severity.

2. AI ACMG module (100.39.0+)

A new card has been added to the AI Shortlist module called AI ACMG module.

Users with a 'Manage AI ACMG' role will have the ability to exclude PP5 and BP6 tags from ACMG classification. These tags rely on external assertions without independent evidence and may introduce bias (Biesecker et al. 2018).

By default, this option is enabled to match current behavior.

• When enabled: PP5 and BP6 tags will not be used in ACMG classification. If either tag is positive (by AI or manual input), a warning message will be displayed in the ACMG section of the variant page.

• When disabled: The tags will be included, and no warning sign will be shown.

3. Pipeline versions

The Pipeline version card, located in the Workbench & Pipeline section of Organization Settings, allows you to set the applied versions for the sample pipeline, case pipeline, DRAGEN, and human reference. You can also configure whether to include reference homozygous genotype calls in cases.

3.1. Sample

Set the organization sample pipeline configurations. These configurations are only applicable for cases starting from FASTQ files.

• Human Ref- Set the used human reference genome build to apply in the pipeline, used for alignment and calling. possible options: GRCh37 and GRCh38.

• Secondary Analysis Pipeline- Set the used secondary analysis pipeline for your organization. This pipeline shall be applicable for all FASTQ samples that are part of the case.

• DRAGEN Version- Set the used DRAGEN version utilized in the Secondary analysis pipeline.

3.2 Variant caller mapping (38.0+)

Select the variant callers enabled for your secondary analysis pipeline.

Each caller is annotated with its sequencing compatibility (WGS, WES), methodology (e.g., CNV read-depth, small variant, SV split-end), and compatibility requirements across sample, DRAGEN, and case pipeline versions. By default, SNV and CNV callers are always enabled. Additional callers, such as SV, SMN, and STR can now be selectively activated to match evolving analysis needs. Changes made through this interface will apply to newly processed cases after changing selected callers.

Enabling variant caller that is not compatible with the selected pipelines is not supported.

Some variant callers may require an assigned PON file to apply properly for WES sequencing type.

3.3 Sample pipeline arguments (38.0+)

View your Sample pipeline arguments for mapping and calling. In order to modify these arguments, you will need to contact technical support.

If the arguments are not set, they will be displayed as ‘N/A’.

3.4 Case

Set the organization case pipeline configuration. This configuration is applicable for all cases in your organization.

Changing the case pipeline will affect the used annotation sources and version for your cases and may impact AI shortlist outcome for your case.

4. Organization DB management (38.0+)

The Organization Database Management feature gives users better visibility and control over the databases (DBs) created and uploaded within their organization. These DBs help improve variant interpretation and are commonly used for filtering known variants (historic or noise) or storing curated findings.

To manage your organization’s databases:

1. Go to Organization Settings

2. Select the Workbench & Pipeline tab

3. Scroll to the new section called Organization DB Management

4.1 Viewing Existing Databases

All users can view a table listing the current databases configured for their organization. If no DBs are present, a message will inform you that none are available.

Table Information Includes:

• DB Name

• DB Type (Historic, Noise, or Curated)

• DB File Name

• Human Reference (e.g., GRCh37, GRCh38)

• Variant Type (SNV or CNV)

• Fields:

  • Historic/Noise: AF, HET, HOM, HEM

  • Curated: Pathogenicity

• Overlap Details (for CNVs only):

  • C: Candidate Overlap

  • A: Annotation DB Overlap

  • Example: C: 70%; A: 70%

  • SNVs will show “N/A” in this column

• AI Shortlist: On/Off

• Active: On/Off

• Last Edited Date

Additional Features:

• Search: Use the search bar to filter the table by DB name

• Download: Click the download icon to export a DB as a VCF file

4.2 Adding a New Database

Users with the Managing Organization DB role can add/edit existing databases within their organization. To create or upload a new Database, please contact your Illumina support representative.

Steps to Add a DB:

1. Click Add New in the Organization DB Management section.

2. Complete the form:

   • Active (On/Off)

   • DB Name

   • DB Type: Historic or Noise

   • Variant Type: SNV or CNV

   • Human Reference: GRCh37 or GRCh38

   • AI Shortlist: On/Off

   • Fields will be auto-populated and not editable

3. Upload the DB file using the file browser.

4. Click Save to finalize. A success message will confirm the addition.

Steps to Edit:

1. Click the Edit button next to the database you want to update.

2. Modify the available fields.

3. Click Save to apply the changes. A confirmation message will appear.

Note: Legacy databases cannot be edited. A notification will indicate if a DB is legacy.

4.3 Audit Logging

All changes (adding or editing a DB) are automatically recorded in the organization’s audit log for transparency and traceability.

5. SV annotation threshold (38.0+)

The Emedgene SV annotation pipeline integrates multiple structural variant databases, including allele frequency sources (e.g., gnomAD SV) and variant/region pathogenicity resources (e.g., ClinVar, ClinGen). Annotation is performed based on defined overlap thresholds tailored to the clinical significance of each database category

Present under Workbench & Pipeline in Organization Settings, this module enables configuration of annotation overlapping thresholds for structural variants with external and internal databases.

5.1 One Side Pathogenic

The One Side Pathogenic setting allows to set a threshold for identifying structural variants classified as pathogenic and likely pathogenic, such as those in ClinGen Pathogenic, ClinVar Pathogenic, DDDSyndromes, and Curate Pathogenic databases. It starts at a default of 0.7, and accepts value between 0 and 1.

5.2 One Side Uncertain

The One Side Uncertain setting allows to set a threshold for identifying structural variants classified as uncertain, such as those in ClinGen VUS, ClinVar VUS, and Curate VUS databases. It starts at a default of 0.7, and accepts value between 0 and 1.

5.3 Two Sided

The Two Sided setting allows to set a threshold for identifying structural variants classified as benign and likely benign, such as those in ClinGen Benign, ClinVar Benign, DECIPHER, DGV, gnomAD SV, 1000 genomes, and Curate Benign databases. It starts at a default of 0.7, and accepts value between 0 and 1.