Українська English

ISSN 2522-9028 (Print)
ISSN 2522-9036 (Online)
DOI: https://doi.org/10.15407/fz

Fiziologichnyi Zhurnal

(English title: Physiological Journal)

is a scientific journal issued by the

Bogomoletz Institute of Physiology
National Academy of Sciences of Ukraine

Editor-in-chief: V.F. Sagach

The journal was founded in 1955 as
1955 – 1977 "Fiziolohichnyi zhurnal" (ISSN 0015 – 3311)
1978 – 1993 "Fiziologicheskii zhurnal" (ISSN 0201 – 8489)
1994 – 2016 "Fiziolohichnyi zhurnal" (ISSN 0201 – 8489)
2017 – "Fiziolohichnyi zhurnal" (ISSN 2522-9028)

Fiziol. Zh. 2025; 71(5S): 75-79

Foundational Models in Biology: When hammering nails with a microscope makes sense

O. Mezhenskyi1, D. Kravchuk1

  1. Bogomoletz Institute of Physiology NASU
DOI: https://doi.org/10.15407/fz71.05.075


Abstract

Foundational models (FMs)—large pre-trained neural architectures—are transforming modern biology by providing universal representations learned from massive, heterogeneous, and often unlabeled datasets. Unlike classical task-specific machine-learning models, FMs can be fine-tuned for genomics, cheminformatics, bioimaging, and physiological signal analysis with minimal amounts of labeled data. This mini-review summarizes key applications of DNABERT, MolBERT, DiffDock, and Segment Anything, highlighting their advantages in accuracy, generalizability, and multimodal integration. We also outline the potential of FMs in physiology and neurophysiology, where they may unify signals from patch-clamp recordings, microelectrode measurements, and calcium imaging into a single analytical framework..

Keywords: : foundational models; machine learning; DNABERT; MolBERT; DiffDock; Segment Anything; bioinformatics; neurophysiology; patch-clamp; image analysis

Full text: (PDF, in English)

References

  1. Greener JG, Kandathil SM, Moffat L, Jones DT. A guide to machine learning for biologists. Nat Rev Mol Cell Biol. 2022;23(1):40-55. CrossRef PubMed
  2. Awais M, Naseer M, Khan S, Anwer RM, Cholakkal H, Shah M, et al. Foundational models defining a new era in vision: A survey and outlook. IEEE Trans Pattern Anal Mach Intell. 2023;47(4):2245-64. CrossRef PubMed
  3. Neidlinger P, Nahhas OSME, Muti HS, Lenz T, Hoffmeister M, Brenner H, et al. Benchmarking foundation models as feature extractors for weakly-supervised computational pathology. arXiv. 2024. Available from: http://arxiv.org/abs/2408.15823 CrossRef PubMed
  4. Tran DH, Meunier M, Cheriet F. Multi-domain learning CNN model for microscopy image classification. arXiv. 2023. Available from: http://arxiv.org/abs/2304.10616
  5. Danuser G. Computer Vision in Cell Biology. Cell. 2011;147(5):973-8. CrossRef PubMed
  6. Kirillov A, Mintun E, Ravi N, Mao H, Rolland C, Gustafson L, et al. Segment Anything. Proc IEEE Int Conf Comput Vis. 2023;3992-4003. CrossRef
  7. Stringer C, Wang T, Michaelos M, Pachitariu M. Cellpose: a generalist algorithm for cellular segmentation. Nat Methods. 2021;18(1):100-6. CrossRef PubMed
  8. Low-dimensional embeddings of high-dimensional data. arXiv. 2025. Available from: https://arxiv.org/html/2508.15929v1
  9. Wang C, Jiang Y, Peng Z, Li C, Bang C, Zhao L, et al. Towards a general-purpose foundation model for fMRI analysis. 2025.
  10. Zhou J, Troyanskaya OG. Predicting effects of noncoding variants with deep learning-based sequence model. Nat Methods. 2015;12(10):931-4. CrossRef PubMed PubMedCentral
  11. Ji Y, Zhou Z, Liu H, Davuluri RV. DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome. Bioinformatics. 2021;37(15):2112-20. CrossRef PubMed PubMedCentral
  12. Zhou Z, Ji Y, Li W, Ramana D, Davuluri V, et al. DNABERT-2: Efficient foundation model and benchmark for multi-species genome. 2024. Available from: https://arxiv.org/pdf/2306.15006v2
    13. Finding the Tree of Life in Evo2. 2025. Available from: https://www.goodfire.ai/research/phylogeny-manifold
  13. Brixi G, Durrant MG, Ku J, Poli M, Brockman G, Chang D, et al. Genome modeling and design across all domains of life with Evo2. bioRxiv. 2025. Available from: https://www.biorxiv.org/content/10.1101/2025.02.18.638918v1 CrossRef
  14. Keyvanpour MR, Shirzad MB. An analysis of QSAR research based on machine learning concepts. Curr Drug Discov Technol. 2021;18(1):17-30. CrossRef PubMed
  15. Graph neural networks for materials science and chemistry. Commun Mater. 2022. Available from: https://www.nature.com/articles/s43246-022-00315-6
  16. Chithrananda S, Grand G, Ramsundar B. ChemBERTa: Large-scale self-supervised pretraining for molecular property prediction. 2020. Available from: https://arxiv.org/pdf/2010.09885
  17. Li J, Jiang X. Mol-BERT: An effective molecular representation with BERT for molecular property prediction. Wirel Commun Mob Comput. 2021;2021:7181815. CrossRef
  18. Wu Z, Ramsundar B, Feinberg EN, Gomes J, Geniesse C, Pappu AS, et al. MoleculeNet: A benchmark for molecular machine learning. arXiv. 2018. Available from: http://arxiv.org/abs/1703.00564 CrossRef PubMed PubMedCentral
  19. Corso G, Stärk H, Jing B, Barzilay R, Jaakkola T. DiffDock: Diffusion steps, twists, and turns for molecular docking. ICLR 2023. Available from: https://arxiv.org/pdf/2210.01776
  20. Wang R, Fang X, Lu Y, Yang CY, Wang S. The PDBbind database: methodologies and updates. J Med Chem. 2005;48(12):4111-9. CrossRef PubMed
  21. Buttenschoen M, Morris GM, Deane CM. PoseBusters: AI-based docking methods fail to generate physically valid poses or generalise to novel sequences. Chem Sci. 2023;15(9):3130-9. CrossRef PubMed PubMedCentral
  22. Johnson AEW, Pollard TJ, Shen L, Lehman LWH, Feng M, Ghassemi M, et al. MIMIC-III, a freely accessible critical care database. Sci Data. 2016;3:160035. CrossRef PubMed PubMedCentral
  23. Clifford GD, Liu C, Moody B, Lehman LH, Silva I, Li Q, et al. AF classification from a short single-lead ECG recording: The PhysioNet/Computing in Cardiology Challenge 2017. Comput Cardiol. 2017;44:1-4. CrossRef PubMed PubMedCentral
© National Academy of Sciences of Ukraine, Bogomoletz Institute of Physiology, 2014-2025.