Property Enhancer - a data efficient multi-objective approach for functional antibody optimization
Tagasovska, N.; Ludwiczak, J.; Loukas, A.; Makowski, E.; Mohammadi-Peyhani, H.; Pooja Mahajan, S.; Zadorozhny, K.; Lee, D.; Goodman, A.; Yao-Yu Lin, J.; Kelly, R.; Hotzel, I.; Bevers, J.; D Souza, T.; Koerber, J. T.; Liang, W.-C.; Lafrance-Vanasse, J.; Chen, Y.; Watkins, A.; Dwyer, H.; Ra, S.; Bonneau, R.; Cho, K.; Gligorijevic, V.
Show abstract
In-silico antibody lead optimization remains challenging due to scarce high-quality data, costly experimental validation, and the need to jointly optimize multiple developability properties. Discovery workflows often rely on high-throughput phage, ribosome or yeast display experiments, which yield large but noisy datasets; as leads emerge, strategies shift to low-throughput assays which are precise, yet unscalable. Deep-learning and language-model approaches are hindered by such limited, unreliable measurements. We introduce Property Enhancer (PropEn), a data-efficient framework for low-data, heterogeneous regimes that can simultaneously optimize multiple antibody properties. PropEn proposes a matching-based augmentation that expands the training data with sequence pairs differing by only a few mutations; within each pair the second sequence improves the target value, providing an implicit optimization signal. Extensive in silico and in vitro tests show 10-39x affinity gains across four targets and nine leads, and enable joint multi-property optimization, positioning PropEn as a scalable, general solution.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- AlphaBind, a Domain-Specific Model to Predict and Optimize Antibody-Antigen Binding Affinity 96%
- Humatch - fast, gene-specific joint humanisation of antibody heavy and light chains 96%
- BioPhi: A platform for antibody design, humanization and humanness evaluation based on natural antibody repertoires and deep learning 95%
Similar papers in this journal
- Protein Engineering with Lightweight Graph Denoising Neural Networks 95%
- DrugHIVE: Target-specific spatial drug design and optimization with a hierarchical generative model 94%
- DENVIS: scalable and high-throughput virtual screening using graph neural networks with atomic and surface protein pocket features 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.