Generating Immune-aware SARS-CoV-2 Spike Proteins for Universal Vaccine Design

Dominic Phillips, Hans-Christof Gasser, Sebestyén Kamp, Aleksander Pałkowski, Lukasz Rabalski, Diego A. Oyarzún, Ajitha Rajan, Javier Antonio Alfaro
Proceedings of the 1st Workshop on Healthcare AI and COVID-19, ICML 2022, PMLR 184:100-116, 2022.

Abstract

Dozens of SARS-CoV-2 vaccines have been approved for public use, yet there remains a risk that the virus evolves to escape vaccine protection. This motivates the development of universal vaccines capable of protecting against current and potentially new strains of the virus. A key challenge is the lack of computational tools to design new viral proteins capable of vaccine escape, which could serve as good targets for the development of universal vaccines. Here, we designed VAE capable of generating SARS-CoV-2 spike proteins with variable immune visibility to the cell-mediated immune response. We compared our model with two simpler generative models; a random-mutator and an 11-gram language model. All three models can generate stable, structurally valid sequences, yet only the VAE model can generate low immunogenicity sequences that interpolate smoothly along the principal variance directions of known natural sequences. This model provides an effective computational tool for the generation of spike protein sequences useful for universal vaccine design. We provide its source code at https://github.com/hcgasser/SpikeVAE.

Cite this Paper

BibTeX
@InProceedings{pmlr-v184-phillips22a, title = {Generating Immune-aware SARS-CoV-2 Spike Proteins for Universal Vaccine Design}, author = {Phillips, Dominic and Gasser, Hans-Christof and Kamp, Sebesty\'en and Pa\l{}kowski, Aleksander and Rabalski, Lukasz and Oyarz\'un, Diego A. and Rajan, Ajitha and Alfaro, Javier Antonio}, booktitle = {Proceedings of the 1st Workshop on Healthcare AI and COVID-19, ICML 2022}, pages = {100--116}, year = {2022}, editor = {Xu, Peng and Zhu, Tingting and Zhu, Pengkai and Clifton, David A. and Belgrave, Danielle and Zhang, Yuanting}, volume = {184}, series = {Proceedings of Machine Learning Research}, month = {22 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v184/phillips22a/phillips22a.pdf}, url = {https://proceedings.mlr.press/v184/phillips22a.html}, abstract = {Dozens of SARS-CoV-2 vaccines have been approved for public use, yet there remains a risk that the virus evolves to escape vaccine protection. This motivates the development of universal vaccines capable of protecting against current and potentially new strains of the virus. A key challenge is the lack of computational tools to design new viral proteins capable of vaccine escape, which could serve as good targets for the development of universal vaccines. Here, we designed VAE capable of generating SARS-CoV-2 spike proteins with variable immune visibility to the cell-mediated immune response. We compared our model with two simpler generative models; a random-mutator and an 11-gram language model. All three models can generate stable, structurally valid sequences, yet only the VAE model can generate low immunogenicity sequences that interpolate smoothly along the principal variance directions of known natural sequences. This model provides an effective computational tool for the generation of spike protein sequences useful for universal vaccine design. We provide its source code at https://github.com/hcgasser/SpikeVAE.} }
Endnote
%0 Conference Paper %T Generating Immune-aware SARS-CoV-2 Spike Proteins for Universal Vaccine Design %A Dominic Phillips %A Hans-Christof Gasser %A Sebestyén Kamp %A Aleksander Pałkowski %A Lukasz Rabalski %A Diego A. Oyarzún %A Ajitha Rajan %A Javier Antonio Alfaro %B Proceedings of the 1st Workshop on Healthcare AI and COVID-19, ICML 2022 %C Proceedings of Machine Learning Research %D 2022 %E Peng Xu %E Tingting Zhu %E Pengkai Zhu %E David A. Clifton %E Danielle Belgrave %E Yuanting Zhang %F pmlr-v184-phillips22a %I PMLR %P 100--116 %U https://proceedings.mlr.press/v184/phillips22a.html %V 184 %X Dozens of SARS-CoV-2 vaccines have been approved for public use, yet there remains a risk that the virus evolves to escape vaccine protection. This motivates the development of universal vaccines capable of protecting against current and potentially new strains of the virus. A key challenge is the lack of computational tools to design new viral proteins capable of vaccine escape, which could serve as good targets for the development of universal vaccines. Here, we designed VAE capable of generating SARS-CoV-2 spike proteins with variable immune visibility to the cell-mediated immune response. We compared our model with two simpler generative models; a random-mutator and an 11-gram language model. All three models can generate stable, structurally valid sequences, yet only the VAE model can generate low immunogenicity sequences that interpolate smoothly along the principal variance directions of known natural sequences. This model provides an effective computational tool for the generation of spike protein sequences useful for universal vaccine design. We provide its source code at https://github.com/hcgasser/SpikeVAE.
APA
Phillips, D., Gasser, H., Kamp, S., Pałkowski, A., Rabalski, L., Oyarzún, D.A., Rajan, A. & Alfaro, J.A.. (2022). Generating Immune-aware SARS-CoV-2 Spike Proteins for Universal Vaccine Design. Proceedings of the 1st Workshop on Healthcare AI and COVID-19, ICML 2022, in Proceedings of Machine Learning Research 184:100-116 Available from https://proceedings.mlr.press/v184/phillips22a.html.

Related Material