Library-Free, Machine Learning-Assisted Antibody Humanization

Antibody humanization can involve laborious library construction and monotonous cycles of protein engineering to design a fully humanized, antibody candidate that is developable and retains the same affinity as the parental antibody.

Twist Biopharma Solutions has developed computational pipelines for the automated generation of humanized variants using a combination of:

Using this in silico humanization platform to rapidly generate humanized antibody candidates from murine, VHH, or chimeric antibody hits, bypasses the need to build and mine an antibody library, cutting down the time to deliver a humanized antibody by more than half.

 

Fully Humanized Affinity Assessment

Using Twist’s rapid humanization pipeline all three murine antibodies were successfully humanized. For a few candidates, the affinity of the top humanized variant was even modestly improved compared to its chimeric control.

Technical Resources

 

Machine Learning Enabled Humanization for More Efficient Antibody Engineering

Case Studies   Biopharma

Download the PDF

Get in Touch With Us

Antibody enthusiasts are standing by and would love to hear more about your discovery needs. Challenging targets welcome!