Tomáš Pluskal Group
CS

Publications

You can find the complete list of our published work in the IOCB publication database. For the full list including pre-prints, refer to the Google Scholar page.

IOCB Publication Database
arrow
Google Scholar
arrow

In our lab, we conduct interdisciplinary research that combines biochemistry, metabolomics, machine learning, and synthetic biology. Below, we selected a few representative publications from our research in each of the aforementioned fields.

Exploring biosynthesis & chemodiversity of natural products

  • We study biosyntheses in selected plant families and the chemodiversity of their natural products.

Computational metabolomics reveals overlooked chemodiversity of alkaloid scaffolds in Piper fimbriulatum.
Damiani, T., Smith, J., Hebra, T., Perkovic, M., Cicak, M., Kadlecova, A., ... & Pluskal, T.
The Plant Journal, 2025.

    Studying Plant Specialized Metabolites Using Computational Metabolomics Strategies.
    Mutabdžija, L., Myoli, A., de Jonge, N. F., Damiani, T., Schmid, R., van der Hooft, J. J., ... & Pluskal, T.
    Plant Functional Genomics: Methods and Protocols, 2024.

    The biosynthetic origin of psychoactive kavalactones in kava.
    Pluskal, T., Torrens-Spence, M. P., Fallon, T. R., De Abreu, A., Shi, C. H., & Weng, J. K.
    Nature Plants, 2019.

    Angiosperm tree of life mapped with literature reports, mined from Wikidata, of the alkaloid scaffolds considered in the present study (i.e., benzylisoquinoline, aporphine, piperolactam, piperidine, seco-benzylisoquinoline). Source: Computational metabolomics reveals overlooked chemodiversity of alkaloid scaffolds in Piper fimbriulatum.

    Computational metabolomics

    • We develop advanced computational tools to enhance the identification and interpretation of mass spectrometry (MS) data.

    Statistical analysis of feature-based molecular networking results from non-targeted metabolomics data.
    Pakkir Shah, A. K., Walter, A., Ottosson, F., Russo, F., Navarro-Diaz, M., Boldt, J., ... & Petras, D.
    Nature Protocols, 2025.

    MassSpecGym: A benchmark for the discovery and identification of molecules.
    Bushuiev, R., Bushuiev, A., de Jonge, N., Young, A., Kretschmer, F., Samusevich, R., ... & Pluskal, T.
    Advances in Neural Information Processing Systems, 2025.

    Emergence of molecular structures from repository-scale self-supervised learning on tandem mass spectra.
    Bushuiev, R., Bushuiev, A., Samusevich, R., Brungs, C., Sivic, J., & Pluskal, T.
    Nature Biotechnology (in press), 2025.

    Efficient generation of open multi-stage fragmentation mass spectral libraries.
    Brungs, C., Schmid, R., Heuckeroth, S., Mazumdar, A., Drexler, M., Šácha, P., ... & Pluskal, T.
    Nature Methods (in press), 2024.

    Reproducible mass spectrometry data processing and compound annotation in MZmine 3.
    Heuckeroth, S., Damiani, T., Smirnov, A., Mokshyna, O., Brungs, C., Korf, A., ... & Pluskal, T.

    Nature Protocols, 2024.

    Integrative analysis of multimodal mass spectrometry data in MZmine 3.
    Schmid, R., Heuckeroth, S., Korf, A., Smirnov, A., Myers, O., Dyrlund, T. S., ... & Pluskal, T.
    Nature Biotechnology, 2023.

    The DreaMS Atlas is built as a three-nearest-neighbor NN-Descent graph (3-NN) based on DreaMS embedding similarities between MS/MS spectra from GeMS-C1, MoNA, and NIST20. Each node includes DreaMS-based property predictions (e.g., druglikeness) and MassIVE metadata (e.g., species). Source: Emergence of molecular structures from repository-scale self-supervised learning on tandem mass spectra.

    Molecular networking is strategy for untargeted MS data clustering, annotation propagation and visualization. Briefly, it organizes untargeted MS data into networks where ions sharing similar MS spectra appear as connected nodes. Source: Reproducible mass spectrometry data processing and compound annotation in MZmine 3.

    Machine learning powered enzyme development

    • We are working towards controlled engineering of novel enzymes with desired enzymatic activity by leveraging state-of-the-art machine learning methods.

    Highly accurate discovery of terpene synthases powered by machine learning reveals functional terpene cyclization in Archaea.
    Samusevich, R., Hebra, T., Bushuiev, R., Bushuiev, A., Čalounová, T., Smrčková, H., ... & Pluskal, T.
    bioRxiv, 2024.

    Revealing data leakage in protein interaction benchmarks.
    Bushuiev, A., Bushuiev, R., Sedlar, J., Pluskal, T., Damborsky, J., Mazurenko, S., & Sivic, J.
    International Conference on Learning Representations, 2024.

    Learning to design protein-protein interactions with enhanced generalization.
    Bushuiev, A., Bushuiev, R., Kouba, P., Filkin, A., Gabrielova, M., Gabriel, M., ... & Sivic, J.
    International Conference on Learning Representations, 2024.

    Machine learning-guided protein engineering.
    Kouba, P., Kohout, P., Haddadi, F., Bushuiev, A., Samusevich, R., Sedlar, J., ... & Mazurenko, S.
    ACS catalysis, 2023.

    Our predictive pipeline outperforms existing methods on tasks of TPS clustering, TPS detection, and TPS substrate classification. Source: Highly accurate discovery of terpene synthases powered by machine learning reveals functional terpene cyclization in Archaea.

    Overview of PPIformer, a SE(3)-equivariant model generalizing across diverse protein-binder variants. Source: Learning to design protein-protein interactions with enhanced generalization.

    Synthetic biology

    • We use synthetic biology methods and develop novel tools to advance the field.

    POMBOX: A Fission Yeast Cloning Toolkit for Molecular and Synthetic Biology.
    T Hebra, H Smrčková, B Elkatmis, M Prevorovsky, T Pluskal.
    ACS Synthetic Biology, 2023. Cover Article.

    Cover page of an ACS Synthetic Biology issue dedicated to our paper introducing POMBOX, a fission yeast cloning toolkit for molecular and synthetic biology.