Anton Bushuiev

anton.bushuiev@cvut.cz | LinkedIn | Bluesky | X | GitHub | Google Scholar | Hugging Face

I am a third-year PhD student at the Czech Technical University in Prague, supervised by Josef Sivic. My primary research interest is developing AI to discover new molecules, such as drugs or enzymes.

I have been working on machine learning methods for designing and optimizing protein binders, as well as for discovering small molecules from nature, with a focus on deciphering mass spectrometry data. Currently, I am combining these two research directions to design enzymes with desired functions. We closely collaborate with my brother, Roman Bushuiev.

I come from Donetsk, Ukraine.

News

Jul 01, 2025 I will be doing a research internship in the Martin Steinegger lab at Seoul National University from July to September.
Apr 22, 2025 Check out an interview with me and my brother on CzechCrunch (in Czech).
Jan 15, 2025 I will be working on the upcoming version of Boltz with the Regina Barzilay group at MIT from January to April (remotely due to a visa delay).

Selected publications

  1. boltzgen.jpeg
    BoltzGen: Toward Universal Binder Design
    Hannes Stark, Felix Faltings*, MinGyu Choi*, Yuxin Xie*, Eunsu Hur*, Timothy O’Donnell*, Anton Bushuiev*, Talip Uçar*, and 30 more authors
    Preprint, 2025
    BoltzGen is a generative model for binder design based on Boltz-2
    DOI Blog Code
  2. proteinttt.gif
    One protein is all you need
    Anton Bushuiev*, Roman Bushuiev*, Olga Pimenova, and 9 more authors
    Preprint, 2025
    ProteinTTT enables customizing protein language models to one protein at a time for enhanced performance on challenging targets
    DOI Code
  3. ppiformer.png
    Learning to design protein-protein interactions with enhanced generalization
    Anton Bushuiev*, Roman Bushuiev*, Petr Kouba, and 8 more authors
    ICLR, 2024
    PPIRef is the first comprehensive dataset of protein-protein interfaces, enabling to train PPIformer for protein-protein interaction design
    DOI Video Code Dataset Poster NEWS
  4. dreams.png
    Self-supervised learning of molecular representations from millions of tandem mass spectra using DreaMS
    Roman Bushuiev*, Anton Bushuiev*, Raman Samusevich, and 3 more authors
    Nature Biotechnology, 2025
    DreaMS is the first foundation model for mass spectrometry
    DOI Code Dataset Tutorials Poster NEWS
  5. massspecgym.png
    MassSpecGym: A benchmark for the discovery and identification of molecules
    Roman Bushuiev, Anton Bushuiev, Niek F. Jonge, and 27 more authors
    NeurIPS Datasets & Benchmarks (Spotlight paper), 2024
    MassSpecGym is the first comprehensive benchmark for molecular discovery with mass spectrometry
    DOI Video Code Dataset Poster NEWS