Convergence of cryo-electron microscopy and artificial intelligence in integrative structural biology: A critical review of advances, synergies, and implications for molecular biophysics and drug discovery.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος:	Convergence of cryo-electron microscopy and artificial intelligence in integrative structural biology: A critical review of advances, synergies, and implications for molecular biophysics and drug discovery.
Συγγραφείς:	Abinawanto; Department of Biology, Faculty of Mathematics and Natural Sciences, Universitas Indonesia, Depok, Indonesia., Sophian A; Food and Drug Investigation Laboratory, The Indonesian Food and Drug Authority (BPOM), Jakarta, Indonesia.
Πηγή:	Journal of microscopy [J Microsc] 2026 Sep; Vol. 303 (3), pp. 235-248. Date of Electronic Publication: 2026 Jul 27.
Τύπος έκδοσης:	Journal Article; Review
Γλώσσα:	English
Στοιχεία περιοδικού:	Publisher: Published for the Royal Microscopical Society by Blackwell Scientific Publications Country of Publication: England NLM ID: 0204522 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1365-2818 (Electronic) Linking ISSN: 00222720 NLM ISO Abbreviation: J Microsc Subsets: MEDLINE
Imprint Name(s):	Original Publication: Oxford, Published for the Royal Microscopical Society by Blackwell Scientific Publications.
Ιατρικοί όροι (MeSH):	Cryoelectron Microscopy/methods , Image Processing, Computer-Assisted/methods , Drug Discovery/methods , Biophysics/methods , Artificial Intelligence*, Electron Microscope Tomography/methods
Περίληψη:	Structural biology has entered a period of rapid methodological change defined by the convergence of cryo-electron microscopy (cryo-EM) and artificial intelligence (AI)-driven structure prediction. Prior reviews have generally addressed cryo-EM instrumentation or AI structure prediction individually, or their combination primarily from a structural biology or drug discovery perspective; a microscopy-centred synthesis tracing this convergence from detector physics and image-processing workflows through validation standards to in situ cryo-electron tomography (cryo-ET), while situating AI tools specifically as inputs to and complements of the cryo-EM workflow, has been comparatively underexplored. This review provides such a synthesis, organised around: (i) the instrumentation, image-processing, and validation advances underlying cryo-EM's resolution gains, including detector physics, contrast transfer function (CTF) estimation, Bayesian and deep-learning-based particle picking and reconstruction, and map-to-model validation and deposition; (ii) cryo-electron tomography and subtomogram averaging (STA) for in situ structural biology, including deep generative approaches to heterogeneity analysis; and (iii) the AlphaFold2, AlphaFold3, and RoseTTAFold All-Atom systems, considered specifically in relation to how they interface with and depend upon cryo-EM data for model building, refinement, and validation. Building on this microscopy-centred account, we propose a three-tier classification framework for selecting between AI-primed, experiment-led, and in situ integrative pipelines, and we summarise performance, transparency, and accessibility differences between leading AI tools. We illustrate these methods with examples from membrane proteins, ion channels, and viral glycoproteins, and briefly note implications for structure-based drug design and for structural biology capacity in resource-limited settings. Throughout, we distinguish between demonstrated capability, exceptional proof-of-principle results, and routine practice, and identify the experimental and computational limitations that constrain each approach. (© 2026 Royal Microscopical Society.)
References:	Nat Methods. 2019 Nov;16(11):1153-1160. (PMID: 31591578) Nature. 2021 Aug;596(7873):590-596. (PMID: 34293799) Nat Commun. 2024 Mar 13;15(1):2265. (PMID: 38480681) Nature. 2021 Aug;596(7873):583-589. (PMID: 34265844) PLoS Comput Biol. 2025 Aug 11;21(8):e1013367. (PMID: 40788932) Elife. 2018 Nov 09;7:. (PMID: 30412051) Nature. 2013 Dec 5;504(7478):107-12. (PMID: 24305160) Int J Mol Sci. 2023 Sep 30;24(19):. (PMID: 37834233) Nat Struct Mol Biol. 2024 Jul;31(7):997-1000. (PMID: 38977902) Science. 2014 Mar 28;343(6178):1443-4. (PMID: 24675944) Science. 2024 Apr 19;384(6693):eadl2528. (PMID: 38452047) Membranes (Basel). 2025 Dec 01;15(12):. (PMID: 41441071) Protein Sci. 2018 Jan;27(1):293-315. (PMID: 29067766) Expert Opin Drug Discov. 2025 Feb;20(2):163-176. (PMID: 39789967) J Struct Biol. 2015 Nov;192(2):216-21. (PMID: 26278980) Front Mol Biosci. 2025 Oct 23;12:1688455. (PMID: 41209584) J Biochem. 2024 Jul 1;176(1):1-10. (PMID: 38498911) Q Rev Biophys. 1988 May;21(2):129-228. (PMID: 3043536) Nat Methods. 2024 Aug;21(8):1525-1536. (PMID: 38459385) Proc Natl Acad Sci U S A. 2021 Jan 12;118(2):. (PMID: 33361332) Nat Methods. 2024 Aug;21(8):1537-1545. (PMID: 39025970) Front Mol Biosci. 2024 Jul 24;11:1429180. (PMID: 39114367) Acta Crystallogr D Struct Biol. 2018 Jun 1;74(Pt 6):531-544. (PMID: 29872004) Nature. 2024 Apr;628(8007):450-457. (PMID: 38408488) Nat Methods. 2021 Feb;18(2):176-185. (PMID: 33542510) Nat Protoc. 2016 Nov;11(11):2054-65. (PMID: 27685097) Nature. 2024 Jun;630(8016):493-500. (PMID: 38718835) Sci Adv. 2025 Sep 26;11(39):eadx4647. (PMID: 40991709) Brief Bioinform. 2025 Nov 1;26(6):. (PMID: 41313605) Nat Rev Drug Discov. 2019 Jan;18(1):59-82. (PMID: 30410121) Commun Biol. 2019 Jun 19;2:218. (PMID: 31240256) Nature. 2023 Oct;622(7981):195-201. (PMID: 37730991) Nat Methods. 2017 Apr;14(4):331-332. (PMID: 28250466)
Contributed Indexing:	Keywords: AlphaFold2; AlphaFold3; cryo‐electron microscopy; cryo‐electron tomography; image processing; integrative structural biology
Entry Date(s):	Date Created: 20260727 Date Completed: 20260902 Latest Revision: 20260904
Update Code:	20260904
PubMed Central ID:	PMC13535891
DOI:	10.1111/jmi.70154
PMID:	42504637
Βάση Δεδομένων:	MEDLINE

Full Text Finder

Περιγραφή
ISSN:	1365-2818
DOI:	10.1111/jmi.70154