Recent advances in antibody optimization based on deep learning methods

Abstract

Antibodies currently comprise the predominant treatment modality for a variety of diseases; therefore, optimizing their properties rapidly and efficiently is an indispensable step in antibody-based drug development. Inspired by the great success of artificial intelligence-based algorithms, especially deep learning-based methods in the field of biology, various computational methods have been introduced into antibody optimization to reduce costs and increase the success rate of lead candidate generation and optimization. Herein, we briefly review recent progress in deep learning-based antibody optimization, focusing on the available datasets and algorithm input data types that are crucial for constructing appropriate deep learning models. Furthermore, we discuss the current challenges and potential solutions for the future development of general-purpose deep learning algorithms in antibody optimization.

Keywords: Antibody optimization; Available dataset; Deep learning; Input data type.

Fig. 1. Number of antibody drugs approved by the United States Food and Drug Administration (US FDA) as of July 2024. Data comes from official information released by the US FDA.

Fig. 2. Illustrative scheme showing that deep learning-based methods can accelerate the development of downstream antibody-based technologies, including the optimized ones for neutralizing toxins and viruses, antibody–drug conjugate (ADC)-targeting tumors, bispecific T-cell engagers (BiTEs) for microenvironment targeting, and engineered immune checkpoint inhibitors. When building deep learning methods, a well-curated dataset is indispensable for training and optimizing the models, whereas sequence information alone or combined with structural details can be used as model inputs.