Streamline Antibody Humanization with Humatch on Vecura

What is Humatch?

Humatch is a specialized tool developed by the Oxford Protein Informatics Group (OPIG) for the rapid, gene-specific humanization of antibody variable-region sequences. By utilizing a combination of Convolutional Neural Network (CNN) classifiers trained on the Observed Antibody Space (OAS) and germline-likeness lookup arrays, it effectively scores and improves the human-likeness of non-human antibodies. It operates without the need for structural data, offering a computationally efficient alternative to traditional structure-based methods.

It helps users iteratively mutate non-human residues toward a target humanness threshold while maintaining essential CDR loops. It is especially useful for researchers in drug discovery and antibody engineering who need to optimize therapeutic candidates for clinical development while ensuring structural integrity.

What can users do with Humatch on Vecura?

With Humatch on Vecura, users can:

• Classify antibody humanness: Quickly assess the human-likeness of VH/VL pairs using specialized CNN classifiers.
• Optimize antibody sequences: Automatically propose framework mutations to humanize non-human sequences while protecting CDRs.
• Perform V-gene identification: Identify the closest human germline V-genes for heavy and light chains.
• Customize engineering constraints: Define specific IMGT positions to fix during the mutation process, giving users granular control over the design outcome.

What the output means

The output provides comprehensive humanness scores for heavy, light, and paired chains, along with a humanized sequence ready for further analysis. It also includes an edit-distance report to track the extent of modifications applied.

This output should be used to support scientific decision making. It does not replace experimental validation.

Why this matters

The clinical success of antibody therapeutics often hinges on minimizing immunogenicity, which is largely driven by the "humanness" of the antibody framework. Humatch streamlines this process by moving beyond simple sequence alignment to a machine-learning-driven approach that recognizes the complex, nuanced patterns of human antibody repertoires.

By enabling high-throughput, automated humanization, Humatch significantly accelerates the design phase of antibody development. This allows researchers to explore more candidates with greater confidence, ultimately shortening the timeline from discovery to clinical validation.

• Developed by: Oxford Protein Informatics Group (OPIG)
• Source: Official GitHub Repository
• Reference: Humatch preprint (bioRxiv 2024)