Skip to main content

Repeat Expansions in Franklin

Written by Support

Overview

Repeat expansions are an important class of genomic variants associated with a growing number of inherited disorders. Unlike conventional sequence variants, their clinical significance depends on the repeat motif, its genomic location, the repeat size, and established disease-specific thresholds.

Repeat expansions are a type of SV in Franklin. Franklin provides the details you need for reviewing and interpreting a repeat expansion variant. For every supported locus (see Supported Repeat Expansions below), this includes the associated gene and condition, the repeat motif and its location, the number of repeats for each allele, and the classification based on the number of repeats. You can filter, sort, and add repeat expansions to the Report similar to other variants.

This article details how disease-associated repeat expansions are annotated and presented in Franklin, how to interpret the information provided during variant review, and which loci are currently supported.

Supported Repeat Expansions

Franklin supports the annotation of repeat expansion variants from output files of the following callers:

  • ExpansionHunter - short-read sequencing

  • TRGT - PacBio HiFi long-read sequencing

The complete list of loci supported for annotation in Franklin can be found here.

How Franklin Displays Repeat Expansion Variants

Displayed Information

For each supported locus Franklin generally shows the following information:

  • Gene and disease

  • Repeat motif

  • Repeat count per allele (as emitted by the caller)

  • Classification: Franklin classifies repeat variant as benign / VUS / pathogenic, according to the repeat range:

    • Benign: Normal allele range

    • VUS: Intermediate range or unknown clinical significance

    • Pathogenic: Disease-causing range

  • Clinical evidence - zygosity, inheritance, disease description, and a curated list of references.

  • Confidence

    • ExpansionHunter: Franklin provides a confidence level - High / Medium / Low - for the number of repeats. It indicates how well the caller's raw signal supports the number of repeats.

      • Three parameters drive the confidence score:

        • Caller filter: The caller's own quality filter must pass.

        • Read support: How many reads directly support the repeat (spanning, flanking, or in-repeat reads).

        • Interval width: How wide the reported confidence interval is relative to the reported number of repeats.

      • High confidence - the caller's filter passed, read support is adequate, and the interval width is low relative to the reported number of repeats.

      • Medium confidence - the call is usable, but either read support is weak or the confidence interval width is high. It is advised to treat the number of repeats as an approximation.

      • Low confidence - read support is weak and the confidence interval width is high, or the reported number of repeats is implausibly large. It is advised to treat with caution and consider orthogonal confirmation.

      • Calls that fail the caller's own quality filter are not included in the case.

    • TRGT: Franklin considers every repeat called by TRGT as High confidence.

The above information is displayed on the repeat expansion variant tile and popup:

If a repeat expansion is called as two repeat lengths, Franklin will display it as two separate variants:

Each of them will display Allele B as the alternative called allele, available from the expanded tile.

Note: Some of the above information may only be available for expansion repeats from a specific supported caller.

Viewing Raw VCF Information

As for other variant types in Franklin, the original VCF line emitted by the caller for each repeat expansion can be viewed directly from the variant tile using the "See VCF" option, available from the variant tile location icon menu. This lets you inspect the raw genotype record - FORMAT fields, INFO tags, allele-level counts and confidence intervals - exactly as they appear in the input file:


This feature is useful for quickly cross-checking the annotated count against the caller's output, or copy-pasting the source line into an external report. For more information on "See VCF", please visit the following help center article.

Support

For the definitive list of loci reported for a specific sequencing method or VCF profile, or if you expect a locus on a case and it is not shown, or to require additional caller or format implementation, please contact our support team ([email protected]).

Did this answer your question?