The release of the HELM Editor by the Pistoia Alliance in 2014 inspired a team at Roche and quattro research to use this technology for the registration of antibodies engineered at Roche’s research sites.
In former years, antibodies have followed nature’s standard format: A set of 2 identical light and heavy chains forms an antibody. In current research projects increasingly complex formats are designed by the scientists. These formats pose the challenge to fully describe their chemical structure.
While the HELM Editor provides the power to efficiently build and describe all kind of biologicals, its canvas doesn’t really scale with large molecules like antibodies. Especially if novel antibodies do not follow nature’s standard and thus the localization of the Cys-Cys bonds within and between domains and chains isn’t straight forward any longer.
To tackle this challenge the team led by Stefan Klostermann (Roche Innovation Center Penzberg) has developed the HELM Antibody Editor (HAbE). Like the HELM Editor that is essentially based on a monomer library, HAbE employs a domain library to describe the functional building blocks of antibodies and other conjugated proteins. Based on this library HAbE decomposes raw protein sequences of even complex antibody formats into their domains and fully analyzes and annotates them. This is supported by a mutation library that enables the identification of known mutations in standard domains.
Using an integrated auto-connector algorithm HAbE also aims to automatically assemble the complete drug molecule into its quaternary structure by finding the correct Cys-Cys bonds. The underlying rules are laid down in a flexible, user customizable syntax describing advanced formats and functional modules like scFV/ scFab. The GUI enables further editing of the molecule to close Cys-Cys bonds e.g. not yet covered by the rule set.
The HELM Editor can be called from within HAbE to edit single Antibody domains based on their monomers, e.g. for coupling reactions like the ones used for ADCs. This integrates the power of the HELM Editor’s flexibility packed into the antibody-optimized framework of HAbE.
During all editing steps the HELM notation is kept up-to-date and is finally one part of the registration together with all detailed annotations generated by HAbE and an automatically generated sketch of the final drug.
The feedback on pre-releases and communication with other pharma companies have already demonstrated an overwhelming interest in HAbE. The whole HAbE team would be happy if this 2nd public application in the HELM family would further increase the impact of HELM as the emerging language for the description of biological macro-molecules.
Stefan gave a presentation on the HELM antibody editor at the Pistoia Alliance evening event in London earlier this year. A video of his presentation is available: