Toward real-world automated antibody design with combinatorial Bayesian optimization

Khan, Asif Ali; Cowen-Rivers, Alexander I.; Grosnit, Antoine; Deik, Derrick-Goh-Xin; Robert, Philippe A.; Greiff, Victor; Smorodina, Eva; Rawat, Puneet; Akbar, Rahmad; Dreczkowski, Kamil; Tutunov, Rasul; Bou-Ammar, Dany; Wang, Jun; Storkey, Amos; Bou-Ammar, Haitham

doi:10.1016/j.crmeth.2022.100374

Cited by 23 publications

(21 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A penalized acquisition function is used to collect batches of points, minimizing non-parallelizable computational effort. Khan et al [16] used a CDRH3 trust region to restrict the search to sequences with favourable developability scores.…”

Section: Related Workmentioning

confidence: 99%

“…However, the sheer number of possible CDRH3 sequences in a combinatorial space makes it infeasible to exhaustively examine any antibody simulation framework [19]. Therefore, we need computational tools to guide our exploration of the protein landscape Recently, Bayesian optimization has demonstrated its efficiency in exploring the sequence design space [16,3]. Bellamy et al [6] compared how noise affects different batched Bayesian optimization techniques and introduced a retest policy to mitigate the effect of noise.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sample-efficient Antibody Design through Protein Language Model for Risk-aware Batch Bayesian Optimization

Wang,

Shi

et al. 2023

Preprint

View full text Add to dashboard Cite

Antibody design is a time-consuming and expensive process that often requires extensive experimentation to identify the best candidates. To address this challenge, we propose an efficient and risk-aware antibody design framework that leverages protein language models (PLMs) and batch Bayesian optimization (BO). Our framework utilizes the generative power of protein language models to predict candidate sequences with higher naturalness and a Bayesian optimization algorithm to iteratively explore the sequence space and identify the most promising candidates. To further improve the efficiency of the search process, we introduce a risk-aware approach that balances exploration and exploitation by incorporating uncertainty estimates into the acquisition function of the Bayesian optimization algorithm. We demonstrate the effectiveness of our approach through experiments on several benchmark datasets, showing that our framework outperforms state-of-the-art methods in terms of both efficiency and quality of the designed sequences. Our framework has the potential to accelerate the discovery of new antibodies and reduce the cost and time required for antibody design.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Sample-efficient Antibody Design through Protein Language Model for Risk-aware Batch Bayesian Optimization

Wang,

Shi

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…4 BO has been successfully applied to many materials science problems. Some recent applications include drug design, 7,8 lithium-ion battery optimisation, 9,10 optical material design, [11][12][13] optimisation of thermoelectric properties 14 and energy predictions of materials. 15,16 It has also been applied to optimise operational conditions such as heating conditions 17 and equipment configurations.…”

Section: Introductionmentioning

confidence: 99%

Bayesian optimisation for efficient material discovery: a mini review

Kumar

Jin

2023

Nanoscale

View full text Add to dashboard Cite

show abstract

“…This semantic gap persists across architectural flavours (for example, generative adversarial networks (GANs) 24 , reinforcement learning 25 , variational autoencoders (VAEs) 26 , graph neural networks (GNNs) 19,27 , flow 28,29 and diffusion models 30 ). However, some works performed property-driven generation through probabilistic reparameterization that directly optimize the input to a property prediction model, for example, gradient-based schemes such as PASITHEA 31 , differentiable scaffolding trees 32 and activation maximization 33 or multi-objective Bayesian optimization 34 that has been applied to peptide inhibitor 35 and antibody design 36 . Still, to our knowledge, existing Transformers either tune task-specific heads (see, for example, refs.…”

mentioning

confidence: 99%

Regression Transformer enables concurrent sequence regression and generation for molecular language modelling

Born

Manica

2023

Nat Mach Intell

View full text Add to dashboard Cite

Despite tremendous progress of generative models in the natural sciences, their controllability remains challenging. One fundamentally missing aspect of molecular or protein generative models is an inductive bias that can reflect continuous properties of interest. To that end, we propose the Regression Transformer (RT), a method that abstracts regression as a conditional sequence modelling problem. This introduces a new direction for multitask language models, seamlessly bridging sequence regression and conditional sequence generation. We demonstrate that, despite using a nominal-scale training objective, the RT matches or surpasses the performance of conventional regression models in property prediction of small molecules, proteins and chemical reactions. Critically, priming the same model with continuous properties yields a competitive conditional generative model that outperforms specialized approaches in a substructure-constrained, property-driven molecule generation benchmark. Our dichotomous approach is facilitated by an alternating training scheme that enables the model to decorate seed sequences on the basis of desired property constraints, for example, to optimize reaction yield. We expect that the RT’s capability to jointly tackle predictive and generative tasks in biochemistry can find applications in property-driven, local exploration of the chemical or protein space. Such multitask approaches will pave the road towards foundation models in materials design.

show abstract

Toward real-world automated antibody design with combinatorial Bayesian optimization

Cited by 23 publications

References 87 publications

Sample-efficient Antibody Design through Protein Language Model for Risk-aware Batch Bayesian Optimization

Sample-efficient Antibody Design through Protein Language Model for Risk-aware Batch Bayesian Optimization

Bayesian optimisation for efficient material discovery: a mini review

Regression Transformer enables concurrent sequence regression and generation for molecular language modelling

Contact Info

Product

Resources

About