AB-Amy: machine learning aided amyloidogenic risk prediction of therapeutic antibody light chains

Yuwei Zhou; Ziru Huang; Yushu Gou; Siqi Liu; Wei Yang; Hongyu Zhang; Anthony Mackitz Dzisoo; Jian Huang

doi:10.1093/abt/tbad007

AB-Amy: machine learning aided amyloidogenic risk prediction of therapeutic antibody light chains

Antib Ther. 2023 Apr 12;6(3):147-156. doi: 10.1093/abt/tbad007. eCollection 2023 Jul.

Authors

Yuwei Zhou¹, Ziru Huang¹, Yushu Gou¹, Siqi Liu¹, Wei Yang², Hongyu Zhang³, Anthony Mackitz Dzisoo⁴, Jian Huang¹

Affiliations

¹ School of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China.
² School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China.
³ Research and Development, Zhanyuan Therapeutics Ltd., Hangzhou, Zhejiang 310000, China.
⁴ Bioinformatics, Data and Medical Reporting, Arcencsus GmbH, Rostock, Mecklenburg-Vorpommern 18055, Germany.

Abstract

Over 120 FDA-approved antibody-based therapeutics are used to treat a variety of diseases.However, many candidates could fail because of unfavorable physicochemical properties. Light-chain amyloidosis is one form of aggregation that can lead to severe safety risks in clinical development. Therefore, screening candidates with a less amyloidosis risk at the early stage can not only save the time and cost of antibody development but also improve the safety of antibody drugs. In this study, based on the dipeptide composition of 742 amyloidogenic and 712 non-amyloidogenic antibody light chains, a support vector machine-based model, AB-Amy, was trained to predict the light-chain amyloidogenic risk. The AUC of AB-Amy reaches 0.9651. The excellent performance of AB-Amy indicates that it can be a useful tool for the in silico evaluation of the light-chain amyloidogenic risk to ensure the safety of antibody therapeutics under clinical development. A web server is freely available at http://i.uestc.edu.cn/AB-Amy/.

Keywords: amyloidosis; developability; prediction; support vector machine (SVM); therapeutic antibody.