Humanization of antibodies using a machine learning approach on large-scale repertoire data
<p><strong>Motivation:</strong> Monoclonal antibody therapeutics are often produced from non-human sources (typically murine), and can therefore generate immunogenic responses in humans. Humanization procedures aim to produce antibody therapeutics that do not elicit an immune respo...
| Main Authors: | , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Oxford University Press
2021
|
| _version_ | 1826307837623533568 |
|---|---|
| author | Marks, C Hummer, AM Chin, M Deane, C |
| author_facet | Marks, C Hummer, AM Chin, M Deane, C |
| author_sort | Marks, C |
| collection | OXFORD |
| description | <p><strong>Motivation:</strong> Monoclonal antibody therapeutics are often produced from non-human sources (typically murine), and can therefore generate immunogenic responses in humans. Humanization procedures aim
to produce antibody therapeutics that do not elicit an immune response and are safe for human use, without
impacting efficacy. Humanization is normally carried out in a largely trial-and-error experimental process.
We have built machine learning classifiers that can discriminate between human and non-human antibody
variable domain sequences using the large amount of repertoire data now available.</p>
<p><strong>Results:</strong> Our classifiers consistently outperform the current best-in-class model for distinguishing human from murine sequences, and our output scores exhibit a negative relationship with the experimental immunogenicity of existing antibody therapeutics. We used our classifiers to develop a novel, computational humanization tool, Hu-mAb, that suggests mutations to an input sequence to reduce its immunogenicity. For a set of therapeutic antibodies with known precursor sequences, the mutations suggested by Hu-mAb
show significant overlap with those deduced experimentally. Hu-mAb is therefore an effective replacement for trial-and-error humanization experiments, producing similar results in a fraction of the time.</p> |
| first_indexed | 2024-03-07T07:09:02Z |
| format | Journal article |
| id | oxford-uuid:4c70ce64-92da-451e-9779-8d54c91840fb |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:09:02Z |
| publishDate | 2021 |
| publisher | Oxford University Press |
| record_format | dspace |
| spelling | oxford-uuid:4c70ce64-92da-451e-9779-8d54c91840fb2022-06-10T08:34:39ZHumanization of antibodies using a machine learning approach on large-scale repertoire dataJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:4c70ce64-92da-451e-9779-8d54c91840fbEnglishSymplectic ElementsOxford University Press2021Marks, CHummer, AMChin, MDeane, C<p><strong>Motivation:</strong> Monoclonal antibody therapeutics are often produced from non-human sources (typically murine), and can therefore generate immunogenic responses in humans. Humanization procedures aim to produce antibody therapeutics that do not elicit an immune response and are safe for human use, without impacting efficacy. Humanization is normally carried out in a largely trial-and-error experimental process. We have built machine learning classifiers that can discriminate between human and non-human antibody variable domain sequences using the large amount of repertoire data now available.</p> <p><strong>Results:</strong> Our classifiers consistently outperform the current best-in-class model for distinguishing human from murine sequences, and our output scores exhibit a negative relationship with the experimental immunogenicity of existing antibody therapeutics. We used our classifiers to develop a novel, computational humanization tool, Hu-mAb, that suggests mutations to an input sequence to reduce its immunogenicity. For a set of therapeutic antibodies with known precursor sequences, the mutations suggested by Hu-mAb show significant overlap with those deduced experimentally. Hu-mAb is therefore an effective replacement for trial-and-error humanization experiments, producing similar results in a fraction of the time.</p> |
| spellingShingle | Marks, C Hummer, AM Chin, M Deane, C Humanization of antibodies using a machine learning approach on large-scale repertoire data |
| title | Humanization of antibodies using a machine learning approach on large-scale repertoire data |
| title_full | Humanization of antibodies using a machine learning approach on large-scale repertoire data |
| title_fullStr | Humanization of antibodies using a machine learning approach on large-scale repertoire data |
| title_full_unstemmed | Humanization of antibodies using a machine learning approach on large-scale repertoire data |
| title_short | Humanization of antibodies using a machine learning approach on large-scale repertoire data |
| title_sort | humanization of antibodies using a machine learning approach on large scale repertoire data |
| work_keys_str_mv | AT marksc humanizationofantibodiesusingamachinelearningapproachonlargescalerepertoiredata AT hummeram humanizationofantibodiesusingamachinelearningapproachonlargescalerepertoiredata AT chinm humanizationofantibodiesusingamachinelearningapproachonlargescalerepertoiredata AT deanec humanizationofantibodiesusingamachinelearningapproachonlargescalerepertoiredata |