Format Rescue

This tool allows users to identify and restore the correct format properties of unknown molecules based solely on their sequence sets.

The workflow consists of six steps.

Sequence Upload

Users begin by uploading the sequence sets. It is critical to ensure that the sequence set for each molecule is complete. Through BioGlyph’s structured input form, users can provide up to three layers of information:

Full sequence set (compulsory)
VH/VL pairings — essential when multiple possible pairings exist within the same sequence set
Biological targets (essential in case corresponding Fvs or building blocks are not in the database already)

Providing this information up front accelerates automated annotation and improves accuracy in Steps 3, 4, and 5.

After uploding sequence with the input form:

Component-Level Annotation

In this step, users review and address the component-level annotations derived from the uploaded sequences. This stage closely parallels the sequence validation workflows in Annotator, Sequence Upload, and Variant Analyzer.

Users can resolve system-generated flags (listed below), including those related to insertions, deletions, and other sequence discrepancies. In addition, users may assign new components such as Linkers and single-component building-block Tags.

If any non-addressable flags arise (i.e., issues that cannot be resolved through sequence-level edits or component assignment), they will be clearly labeled for user review.

Assembly of Component Connections and Format Reconstruction

Steps 3 and 4 focus on establishing the non-peptide-bond connections between components, represented through adjacency within BioGlyph’s building-block by the object base language. At this stage, users see the system assemble Building Blocks composed of one or more components — including Fabs, xFabs, scFabs, scFvs, and both homo- and hetero-Fc.

For Building Blocks that contain two protein chains (e.g., Fab, xFab, and Homo- or Hetero-Fc), the assembly process automatically advances the reconstruction of the overall format.

Building Block and Target Association

BioGlyph assigns biological targets to each Building Block. If target information is not provided in the input file, BioGlyph automatically searches its database for sequence-target matches. If a positive match is found: the target is assigned automatically.

If no match is identified: the target remains unassigned

. Users can then optionally review and confirm the assignments or manually associate a target to any Building Block based on additional knowledge or external data sources.

Import of Rescued Molecules

After completing the review, users can import the rescued molecules into the BioGlyph database. Partial rescues are supported.

For example: If not all Building Blocks in a sequence set are identified and the full format cannot be reconstructed, the validated Building Blocks will still be imported.

If one or more targets cannot be assigned, the corresponding Building Blocks will be registered with target status = "unknown".

This ensures that all verified information is captured in the system, even when full annotation is not yet possible

Molecules Page