Skip to content
DraconiticResearch
Search Sources API
201,683 studies31 sourcesUpdated 56 min ago
Peer-reviewedCC BYNature Biotechnology

Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles

  • Advaith Maddipatla1
  • Nadav Bojan Sellam1
  • Meital Bojan1
  • Volodymyr Masalitin1
  • Sanketh Vedula2,3
  • Paul Schanda4
  • Ailie Marx5,6
  • Alex M. Bronstein7,8
8 affiliations
  1. 1Institute of Science and Technology Austria (ISTA), Klosterneuburg, Austria
  2. 2Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, MA, USA. [email protected]
  3. 3Princeton University, Princeton, NJ, USA. [email protected]
  4. 4Institute of Science and Technology Austria (ISTA), Klosterneuburg, Austria. [email protected]
  5. 5MIGAL - Galilee Research Institute, Kiryat Shmona, Israel. [email protected]
  6. 6Tel Hai - University of Kyriat Shmona in the Galilee, Upper Galilee, Israel. [email protected]
  7. 7Institute of Science and Technology Austria (ISTA), Klosterneuburg, Austria. [email protected]
  8. 8Technion - Israel Institute of Technology, Haifa, Israel. [email protected]
Open at sourcePDFDOICite

Abstract

AlphaFold3 predicts highly accurate protein structures from sequence but tends to collapse to a single dominant conformation, even when the underlying structure is inherently heterogeneous. Moreover, its predictions are oblivious to experimental conditions that can alter local sequence conformation. In this work, we show that AlphaFold3 can be guided to match data obtained by nuclear magnetic resonance (NMR) spectroscopy, X-ray crystallography and cryogenic electron microscopy (cryo-EM) experiments and combinations thereof. Our approach can also incorporate data that explicitly report on dynamics, such as site-resolved order parameters. We demonstrate that this methodology generates compact structural ensembles whose ensemble-averaged observables agree with experiment, with fewer distance restraint violations than traditionally resolved NMR structures and with unmodeled alternate conformations uncovered in electron density. This methodology paves the way for experimentally aware predictive models that generate structural ensembles consistent with the measurements, potentially over multiple modalities, and that can be further refined toward thermodynamically grounded ensembles by incorporating energetics.

Related

  1. Bioinformatic prediction, qCLASH, and AlphaFold3 reveal non-overlapping layers of the miR-214-3p–mRNA interactome
    Jia Yao, Bingqiao Huang, Chenyu Chang +5
    Frontiers in geneticsPeer-reviewedCC BYFull textGeneral biochemistryEurope PMC
  2. Structure bioinformatics of eight human ATP synthase F₀ subunits and their AlphaFold3-predicted reduced-hydrophobicity QTY analogues
    Ziyu Wang, Edward Chen, Shuguang Zhang
    PLOS OnePeer-reviewedCC BYFull textMolecular biology & gene regulationPLOSEurope PMC
  3. Structural thermodynamics by cryo-EM resolves minute energy differences in heterogeneous protein complexes.
    L. Boyens-Thiele, M. Bicanin, K. Engholm-Keller +5
    bioRxivPreprintCC BY-NC-NDFull textBiophysicsbioRxiv
  4. The Virtual Tissues foundation model resolves spatial proteomics across scales
    Johann Wenckstern, Eeshaan Jain, Benedikt von Querfurth +9
    NaturePeer-reviewedCC BY21 citedMolecular biology & gene regulationOpenAlex
  5. Barbell reveals and resolves demultiplexing and trimming issues in Nanopore data
    Rick Beeloo, Ragnar Groot Koerkamp, Xiu Jia +5
    BioinformaticsPeer-reviewedCC BY9 citedMolecular biology & gene regulationOpenAlex

Cite

BibTeXRIS
BibTeX
@article{maddipatla2026experimentguided,
  title = {{Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles}},
  author = {Advaith Maddipatla and Nadav Bojan Sellam and Meital Bojan and Volodymyr Masalitin and Sanketh Vedula and Paul Schanda and Ailie Marx and Alex M. Bronstein},
  journal = {Nature Biotechnology},
  year = {2026},
  month = jun,
  doi = {10.1038/s41587-026-03166-5},
  url = {https://doi.org/10.1038/s41587-026-03166-5},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Experiment-guided AlphaFold3 resolves measurement-consistent protein ensembles
AU  - Advaith Maddipatla
AU  - Nadav Bojan Sellam
AU  - Meital Bojan
AU  - Volodymyr Masalitin
AU  - Sanketh Vedula
AU  - Paul Schanda
AU  - Ailie Marx
AU  - Alex M. Bronstein
PY  - 2026
DA  - 2026/06/29
JO  - Nature Biotechnology
DO  - 10.1038/s41587-026-03166-5
UR  - https://doi.org/10.1038/s41587-026-03166-5
AB  - AlphaFold3 predicts highly accurate protein structures from sequence but tends to collapse to a single dominant conformation, even when the underlying structure is inherently heterogeneous. Moreover, its predictions are oblivious to experimental conditions that can alter local sequence conformation. In this work, we show that AlphaFold3 can be guided to match data obtained by nuclear magnetic resonance (NMR) spectroscopy, X-ray crystallography and cryogenic electron microscopy (cryo-EM) experiments and combinations thereof. Our approach can also incorporate data that explicitly report on dynamics, such as site-resolved order parameters. We demonstrate that this methodology generates compact structural ensembles whose ensemble-averaged observables agree with experiment, with fewer distance restraint violations than traditionally resolved NMR structures and with unmodeled alternate conformations uncovered in electron density. This methodology paves the way for experimentally aware predictive models that generate structural ensembles consistent with the measurements, potentially over multiple modalities, and that can be further refined toward thermodynamically grounded ensembles by incorporating energetics.
SN  - 0733-222X
SN  - 1087-0156
SN  - 1546-1696
SN  - 2331-3684
ER  -