New in Emedgene V35.0 (May 22nd 2024)

Release highlights for both workbench 35.0 and pipeline 35.0:

• New AI capabilities: Sort by AI rank & Phenomeld score enhances use of AI models in analysis workflow.

• New gene list workflows: Apply multiple gene lists during and after case creation, with full traceability.

• Cases page gets two most frequently requested features to improve workload management: assign case to another user, & quality summary.

• CNV interpretation improvements: New visualization for target counts bigwig, sort by size, differentiate between variant callers for the application of noise DBs and more.

• New customer self-serve capabilities: Batch upload to Curate from the UI with validation, easy move/add of gene lists to organizations, new help center.

• Enhanced flexibility for lab implementations: Lab defined region of interest and QC per test type, more QC metrics with DRAGEN reports, custom BED annotations and filtering, efficient API queries.

• Expanding applications supported on Emedgene with a new haplotype variant and DRAGEN JSON ingestion support.

Emedgene customers can select their preferred version out of any of the past 5 releases. Customers on v30.0 and below should select an upgrade path at this time.

Updates

AI | Improved AI outputs for use in the interpretation workflow. AI rank and Phenomeld scores available and sortable in analysis tools.

Analysis Tools | Sort by AI rank

AI rank is now available in a sortable, exportable column in the analysis tools.

Emedgene does not provide the AI score as it can bias the analysis; whereas AI rank provides an unbiased method to utilize AI in the interpretation workflow.

• If two variants share an identical score they will have the same rank in the column.

• A reanalysis will update the AI rank values.

( i ) Requires workbench & pipeline update. Old cases can be rerun to retrieve the rank.

Analysis Tools | Sort by Phenomeld score

Phenomeld score is now available in a sortable, exportable column in the analysis tools, in addition to the available filter. Range is 0-2.

( i ) Requires workbench & pipeline update.

As a reminder, Phenomeld is a proprietary phenotypic match algorithm that estimates the concordance between patient phenotypes and an associated disease. It is an important component in our XAI model.

Phenomeld achieves an AUC of 0.94, and uses ensemble machine learning, which combines several phenotypic match models in order to improve model performance.

The Phenomeld model was validated with two data sets.

Utilizing an internal dataset of 314 well characterized cases, Phenomeld was compared to the open source phen2gene model, as well as to the previous Emedgene Phenomatch model, outperforming both.

A second validation used a set of 4648 ClinVar cases, and compared Phenomeld with phen2gene, where Phenomeld demonstrated improved results.

Apply multiple gene lists at and after case creation

This version improves support for labs that regularly apply multiple gene lists per case.

Add New Case | Apply multiple gene list or add genes to a gene list at case creation, from the UI

When applying a gene list at case creation, you can now create a new gene list comprised of multiple gene lists, or add genes to an existing gene list. The new ‘merged’ gene list will carry the same behavior as any other gene list on the platform.

Choose create a new gene list during add new case. Then, search for a single or multiple gene lists, or add genes to an existing gene list:

( i ) Requires a workbench update.

Analysis Tools | Dynamic gene list and search application to preset filters, with full traceability.

Customers can now dynamically couple a gene list, multiple gene lists, or any other search with their preset filters, and review the restricted subset of preset filters with full traceability. This enhancement addresses the need for users to track the reviewing of different gene lists in their case.

To increase flexibility of analysis, this feature is enabled using the search bar, which is now located above the preset filters.

Simply search for a gene list, multiple gene lists or any other supported search function.

By clicking on the 3 vertical dots, you can save your search as a custom preset filter. If you do not save a custom preset filter, you can still perform the restricted analysis, however it will not be saved, and you will not be able to populate it to a report.

You can then review the case according to any of your saved custom preset filters.

Custom preset filters are saved per case, have a full activity trail and can be automatically populated to a report.

( i ) Requires a workbench update.

Cases Page | Assign case to other user & quality summary for improved workload management.

The cases page is getting two of the most requested updates in this version.

Assign cases to users for improved workload management.

Lab directors can now assign cases to users, streamlining workload management within the lab. Previously only users could subscribe to a case.

( i ) Requires a workbench update.

Cases page now displays the case quality summary.

The case quality summary has been added as a column in the cases page. This quick view of case quality will improve workload allocation within the lab. The column is sortable and can also be filtered on.

Hovering on the quality icon will provide a quick summary of findings, indicating failure for case or sample quality, and also pedigree and gene coverage validation.

( i ) Requires a pipeline & workbench update.

CNV Interpretation Improvements | Added target counts BigWig, sort by CNV size & more.

This version adds multiple features aimed at improving CNV interpretation capabilities, an effort that will continue in subsequent versions.

• New visualization track = the target counts BigWig is now supported, in addition to the tangent normalized BigWig. This is a BigWig representation of the target counts bins. This new track is available from the UI, batch or API for customers starting from both FASTQ or VCF. ( i ) This feature requires both a pipeline and a workbench update.

• Analysis tools – new column for variant length (kb), can be sorted, exported and populated to the report. ( i ) This feature requires a workbench update.
• Analysis tools – New filtering capability by calling methodology. While this new feature applies to any variant caller used by customers, it can be used to create filters for CNVs called by the DRAGEN read-depth caller separately from the DRAGEN SV caller, so that customers can apply separate noise databases to each. The new filter appears under the Quality filters. Possible calling methodologies:

  • Small variant caller

  • CNV read-depth caller

  • Forced Genotyping

  • Star allele caller

  • STR repeat expansion caller

  • SV split-end caller

  • Targeted caller

  • Unknown

( i ) Requires a pipeline & workbench update.

• Annotation is now enabled for CNVs larger than 20M bp. ( i ) Requires pipeline & workbench update

• Visualization | Added the copy number value to CNV variants in the proband VCF track. ( i ) Requires pipeline & workbench update.

• Visualization | Improved performance for the embedded IGV enables quickly moving between variants when multiple new visualization tracks are selected. ( i ) Requires a workbench update.

Self-Serve: Batch upload to Curate with validation, easy move/create gene lists, new help center.

We continue to enable customers to control their Emedgene accounts, and to reduce their reliance on Illumina support teams.

Batch upload to Curate from the UI, with validation.

Customers can now batch add variants to Curate via the UI, utilizing a simple CSV template. Emedgene software will perform a validation on the upload to ensure compatibility and quality in customer Curate databases.

Simply switch to batch mode in the Add new pane in Curate to download an example CSV file and follow the link for more instructions. Note the size limitation for each upload is 5,000 variants. Multiple concurrent uploads are allowed.

After upload, a ‘toaster’ element will appear at the bottom left of Curate. While the toaster can be closed and the upload will continue, the report on the variants uploaded and any that might have failed or partially failed validation can only be accessed through the toaster.

( i ) This feature requires a workbench update.

Easy add/move gene lists in Management and Add New Case.

It’s easier than ever to create new gene lists in the Management page or Add New Case flow.

• When switching to batch mode, both gene symbols and NCBI IDs are accepted.

• When pasting a list of NCBI IDs or gene symbols from a CSV file, whether downloaded from an Emedgene staging organization or other, no further manipulation needed and the list is accepted as is.

( i ) This feature requires a workbench update.

We have a new help center!

We’ve made it easier to find and use our help center by moving it to help.emg.illumina.com no matter where you are located around the world. Previous local versions had limitations with internal linking, these are now fully removed. The new help center can be accessed at this URL regardless of workbench or pipeline versions.

Enhanced flexibility for lab implementations

This version adds multiple features that enhance flexibility to implement lab standard operating procedures.

• Define a region of interest and QC BED for any test type, previously only customization for test type ‘Custom Panel’ was enabled.

  • The region of interest BED, whether default or custom, determines the billing.

    • If the number of variants annotated is <55K, billing is for a panel.

    • If the number of variants annotated is 55K-300K billing is for an exome.

    • If the number of variants is greater than 300K billing is for a genome.

• If a reflex to a wider backbone is needed, the test can be edited and a different BED selected.

• QC BEDs can be uploaded separately.

The Region of Interest and QC BED files can be defined from the Add New Case flow in the UI batch upload, and API.

( i ) Requires a pipeline & workbench update.

• Many new QC metrics are now available in the Lab Tab through a DRAGEN report integration. If your DRAGEN report is available, a link will be added below the sample name. This feature is available for customers starting from FASTQ or VCF from API (See supported DRAGEN outputs below for details).

( i ) Requires a pipeline & workbench update.

• Efficient API queries are here! When fetching details of variant(s) and case(s) via the API, users can now define which values to return, rather than returning all values associated with the query. Each query will return default values, and in addition the fields requested.

( i ) Requires a workbench update.

• Support for custom BED annotations & filtering

Customers can now annotate their cases with a BED file, and filter on the variants based on the BED annotation. This feature requires support assistance for implementation.

( i ) Requires a pipeline & workbench update.

• Push genes of insufficient coverage to a report

Customers can push lists of genes with insufficient coverage at various thresholds (<=0x, <=5x, <=10x, <=20x; or % of BasesGt20x <=20%, <=50%, <=80%, <=90%) and details (transcript, exons affected, position etc) to a report.

( i ) Requires a workbench update

New haplotype support for DRAGEN star allele JSON.

This version expands the possible applications that can be interpreted and reported on in Emedgene to include haplotype driven applications.

This version expands the possible applications that can be interpreted and reported on in Emedgene to include haplotype driven applications.

• Ingest and annotate the DRAGEN star allele caller JSON file (*.targeted.json) including targeted caller CYP2D6 and CYP2B6 outputs as haplotype variants.

• New haplotype variant page displays haplotype, genes affected and quality parameters from DRAGEN.

• Haplotype variants can be interpreted and pushed into a report.

• Report pre-processing capabilities can be utilized to produce PGx reporting by integrating the customer’s own metabolizer status-drug database.

( i ) Requires a pipeline & workbench update.

Additional improvements to workflows and pipeline

• gnomAD PLI scores were added to export csv. ( i ) Requires a workbench update.

• ClinVar SV, ClinGen and MitoMap annotations were added to the monthly update automation infrastructure, and will be updated on a monthly basis going forward.

• DRAGEN pedigree pipeline, which improves accuracy of de novo calling for WGS, is now available for customers starting from FASTQ. ( i ) Requires pipeline & workbench update.

  • Only for ICA customers with input FASTQs in ICA, who are only running WGS in their organization.

  • This feature has always been available for customers who run their own DRAGEN pipeline.

  • From the customer perspective, no experience with ICA needed and this behaves just like the integrated Emedgene DRAGEN pipeline, fully and seamlessly executed for customers.

  • Billing for DRAGEN runs is on ICA with iCredits. On Emedgene, please purchase the Genome Equivalents VCF SKU.

• Customers starting from FASTQ will now run the DRAGEN 4.2 CNV-SV merged file by default. ( i ) Requires a pipeline update.

• This version supports ingestion of outputs from Illumina’s newly released Trusight Whole Genome IVD workflow. ( i ) Requires a pipeline & workbench update.

• This version supports ingestion of VCFs from Illumina’s CFTR IVD workflow. ( i ) Requires a pipeline and a workbench update.

• This version supports ingestion of small variant calling VCFs from the Illumina amplicon pipeline. ( i ) Requires a pipeline & workbench update.

• Supported DRAGEN 4.2 outputs in V35:

Limitations:

• Add New Case | Input file path cannot contain spaces or parenthesis.

• Add New Case | API | Applying multiple panels has a limit of 10,000 genes in total but no error message.

• Add New Case | For customers starting from Joint gVCF, please make sure the proband is first in order to see the correct insufficient region calculation.

• Pipeline | The DRAGEN pedigree pipeline for customers starting from FASTQ is only available for customers who have their FASTQ files stored on ICA, and only run WGS in their organization.

• Edit Case | Reanalysis | When applying a new gene list the original gene list is not displayed in activity.

• Lab Tab | Insufficient coverage export will not work via UI or API if an included gene does not have a start or end position in NCBI.

• AI Shortlist | Will not prioritize chr17:43124027 ACT>A in the BRCA gene which contains >600K links to publications, due to limitations on annotation file size. This limitation will be removed in upcoming patch releases V34.4 and V35.1.

• Analysis Tools | Multi-Select | Does not have an aggregated activity report.

• Analysis Tools | Manually Added Variants | Cannot be sorted on non-applicable column types e.g. AI rank.

• Analysis Tools | Manually Added Variants | STRs | Format is not aligned with format of STRs on the software, e.g. missing variant length.

• Variant Page | Population data | ‘af_max_any_pop’ utilizes rounding to 5 digits.

• Curate | When exporting a variant from Analyze to Curate, users may sometimes see an erroneous error message ‘None is not allowed for pathogenicity’. The actual error is an unsupported variant type.

• Curate | Transcript versions will not be identical in variants created in Curate vs variants created in Analyze. Future enhancement will align transcript selection logic.

• Curate | Discrepancy between Analyze and Curate HGVS parser may be experienced.

• Network | GRCh37<-->GRCh38 Liftover not available for older components of Network infrastructure, as a result, [Variant Page | Clinical Significance | Networks Classified] may remain erroneously empty while [Variant page | Related cases section] shows relevant information. Same gap for manually classified variants.

• Export to excel is limited to 32KB in size, which may prevent exports with very large CNVs.

• Webhooks | Cannot be trigger on internal software statuses such as ‘In Progress’ ‘Reanalysis’.

• Organization Settings | BED upload | Validation on the UI component does not check that all lines in the BED contain the same number of columns.

Fixed Issues:

• General | Users on Illumina clouds will now get a warning when they are logged out due to their IAM settings. The warning will prompt for a refresh, preventing loss of work on the Emedgene software. This issue was fixed in v34.2 patch and is re-documented with this release.

• General | When there is a mismatch of the number of variants in the VCF and the number of variants in the case, a prominent red warning banner appears on the software. Prior to V35 this banner was organization-wide, flashing up for all users even with unaffected cases. Starting with this version, the warning will be case specific.

• Candidate | Evidence Graph | Updated literature links to LitVar2.

• Variant Page | Fixed an issue where in some cases OMIM inheritance modes were incorrectly displaying CGD values.

• Analysis Tools | Fixed an issue that caused PLI filters relying on a ‘lof_intolerant_probability’ query to return partial results.

• Curate | Fixed an issue causing display of Related Cases component for the previously displayed variant.

Known Issues:

• gnomAD 4.1 fix will be supported in V34.5/V35.1. All gnomAD 4.0 links from the UI are directed to 4.1 data, as there is no way to link directly to 4.0 data. This creates a data<-->link discrepancy as Emedgene annotation data is still on gnomAD 4.0.

• Add New Case | Flow to run case as a singleton and add family members in a rerun is temporarily not working for cases starting from FASTQ. Targeted fix for V34.4 & V35.1.

• Add New Case | API | When sending due date please use UTC time, customer time zone is not taken into account with API, only through the UI.

• Edit Case | Reanalysis | If HPO terms were updated between analyses, the reanalysis will not automatically map previous HPO terms to new ones.

• Cases Page | Illumina Clouds | Users that have been removed from workgroups in IAM can still be added as participants to a case. They will not have access to the software, and there is no security/access risk.

• Candidates Page | Compound het SNV-CNV variants will not display the automated CNV classification. Workaround – view variants from analysis table.

• Lab Tab | Average coverage calculation in the UI can miscalculated and over 100% in rare cases. Targeted fix for V34.5 and V35.1.

• Lab Tab | UI issue makes it difficult to identify which sample is selected for viewing quality metrics. Fix planned for V34.5/V35.1.

• Analysis Tools | ‘Last’ button on pagination does not work.

• Variant Page | Rare lag in variant interpretation load time can cause text to be mistakenly overwritten.

• Variant Page | Visualization | Simple/Advanced selectors will not work for locally uploaded BAM files.

• Curate | Searching for a variant causes the related cases to disappear even when search is removed. Work around is to refresh.

• Curate | Orphanet links no longer work due to an Orphanet link structure update.

• Finalized Case | When user enters a variant in a case in finalized status, variant will be marked as viewed. Fix planned for V35.1.

• Activity | Editing interpretation paragraph yields an erroneous activity labeled reanalysis.

• API | Marking over 5000 variants as viewed will not result in an error message, although this is the database limitation for saving viewed variants.

• Dashboard | Diagnostic Yield includes Uncertain as Resolved.