Academic Journal
Προβολή σε EDS
Συνδεδεμένο Πλήρες Κείμενο

Multivalent ion binding site identification with structure-based deep learning.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Multivalent ion binding site identification with structure-based deep learning.
Συγγραφείς: Kozlovskii I; Constructor University Bremen gGmbH, Bremen, Germany.; Constructor Labs, Bremen, Germany., Popov P; Constructor University Bremen gGmbH, Bremen, Germany. ppopov@constructor.university.; Constructor Labs, Bremen, Germany. ppopov@constructor.university.
Πηγή: Communications biology [Commun Biol] 2026 Jul 27; Vol. 9 (1). Date of Electronic Publication: 2026 Jul 27.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: Nature Publishing Group UK Country of Publication: England NLM ID: 101719179 Publication Model: Electronic Cited Medium: Internet ISSN: 2399-3642 (Electronic) Linking ISSN: 23993642 NLM ISO Abbreviation: Commun Biol Subsets: MEDLINE
Imprint Name(s): Original Publication: London, United Kingdom : Nature Publishing Group UK, [2018]-
Ιατρικοί όροι (MeSH): Proteins*/chemistry , Proteins*/metabolism , Ions*/metabolism , Ions*/chemistry , Deep Learning*, Binding Sites ; Protein Binding ; Protein Conformation ; Models, Molecular ; Convolutional Neural Networks
Περίληψη: Protein-ion interactions are essential for many cellular processes, including enzymatic catalysis, signaling, and allosteric regulation. However, mapping ion-binding sites experimentally remains labor-intensive and expensive. Here, we present BiteNetI, a structure-based deep learning model that uses 3D convolutional neural networks to simultaneously localize ion-binding centers and predict binding residues for 14 biologically relevant ions. Trained on a carefully curated dataset of over 10,000 high-resolution protein-ion complexes, in which near-identical binding sites are consistently annotated by transferring ions between homologous structures, BiteNetI shows strong generalization ability across diverse ions within a unified multitask architecture. On two different test benchmarks, BiteNetI achieves state-of-the-art performance compared to existing ion-binding predictors as well as to a more general method, AlphaFold3, when used to predict the entire structure of protein bound to ions. Finally, for physiologically relevant ions such as Ca2+, Na+ and K+, BiteNetI achieves two- to three-fold improvement in accuracy.
(© 2026. The Author(s).)
Competing Interests: Competing interests: The authors declare no competing interests.
References: Commun Biol. 2020 Oct 27;3(1):618. (PMID: 33110179)
Chem Rev. 2014 Jan 8;114(1):538-56. (PMID: 24040963)
Curr Protoc Bioinformatics. 2016 Jun 20;54:5.6.1-5.6.37. (PMID: 27322406)
Science. 2012 Jul 13;337(6091):232-6. (PMID: 22798613)
Mol Biol Evol. 2013 Apr;30(4):772-80. (PMID: 23329690)
J Chem Inf Model. 2022 Jul 11;62(13):3157-3168. (PMID: 35749367)
Commun Biol. 2026 Jul 27;9(1):. (PMID: 42547519)
Nucleic Acids Res. 2000 Jan 1;28(1):235-42. (PMID: 10592235)
Nat Commun. 2023 May 11;14(1):2713. (PMID: 37169763)
Q Rev Biophys. 2025 Mar 12;58:e12. (PMID: 40071518)
Proteins. 2014 Feb;82 Suppl 2:154-63. (PMID: 24339001)
Nucleic Acids Res. 2005 Apr 22;33(7):2302-9. (PMID: 15849316)
Bioinformatics. 2010 May 1;26(9):1160-8. (PMID: 20305268)
Biometals. 2001 Sep-Dec;14(3-4):271-313. (PMID: 11831461)
RSC Adv. 2019 Dec 24;10(1):215-227. (PMID: 35492549)
Nucleic Acids Res. 2011 Jul;39(Web Server issue):W288-92. (PMID: 21576237)
Physiol Rev. 2015 Jan;95(1):1-46. (PMID: 25540137)
IEEE/ACM Trans Comput Biol Bioinform. 2013 Jul-Aug;10(4):994-1008. (PMID: 24334392)
Bioinformatics. 2024 Jan 2;40(1):. (PMID: 38175787)
FASEB J. 2015 Oct;29(10):4174-88. (PMID: 26116702)
Brief Bioinform. 2022 Nov 19;23(6):. (PMID: 36274238)
Bioinformatics. 2016 Nov 1;32(21):3260-3269. (PMID: 27378301)
Nature. 2021 Aug;596(7873):583-589. (PMID: 34265844)
Nucleic Acids Res. 2022 Jan 7;50(D1):D439-D444. (PMID: 34791371)
J Biol Inorg Chem. 2008 Nov;13(8):1205-18. (PMID: 18604568)
Nature. 2024 Jun;630(8016):493-500. (PMID: 38718835)
BMC Bioinformatics. 2016 Feb 10;17:81. (PMID: 26864881)
Bioinformatics. 2015 Apr 15;31(8):1322-4. (PMID: 25505090)
Philos Trans R Soc Lond B Biol Sci. 2018 Jun 19;373(1749):. (PMID: 29735728)
Bioinformatics. 2025 Aug 2;41(8):. (PMID: 40742755)
J Med Chem. 2019 Jan 10;62(1):88-127. (PMID: 30106578)
Anal Biochem. 2019 Feb 1;566:75-88. (PMID: 30414728)
J Chem Inf Model. 2016 Dec 27;56(12):2287-2291. (PMID: 27976886)
Bioinformatics. 2013 Oct 15;29(20):2588-95. (PMID: 23975762)
Chem Soc Rev. 2021 Aug 21;50(16):9104-9120. (PMID: 34184009)
NAR Genom Bioinform. 2021 Nov 26;3(4):lqab111. (PMID: 34859211)
J Mol Biol. 1993 Nov 20;234(2):463-82. (PMID: 8230226)
Proc Natl Acad Sci U S A. 2020 Sep 1;117(35):21711-21722. (PMID: 32817431)
Pharmacol Rev. 2019 Oct;71(4):571-595. (PMID: 31551350)
Nucleic Acids Res. 2021 Jan 8;49(D1):D335-D343. (PMID: 33270898)
Biochemistry. 2017 Feb 21;56(7):944-956. (PMID: 28121131)
J Chem Inf Model. 2021 Aug 23;61(8):3814-3823. (PMID: 34292750)
Nat Chem Biol. 2023 May;19(5):548-555. (PMID: 36593274)
Trends Biochem Sci. 2014 May;39(5):233-44. (PMID: 24767681)
Brief Bioinform. 2022 Sep 20;23(5):. (PMID: 35595534)
Nat Rev Drug Discov. 2017 Dec;16(12):829-842. (PMID: 29075003)
Substance Nomenclature: 0 (Proteins)
0 (Ions)
Entry Date(s): Date Created: 20260803 Date Completed: 20260804 Latest Revision: 20260813
Update Code: 20260814
PubMed Central ID: PMC13434133
DOI: 10.1038/s42003-026-10659-1
PMID: 42547519
Βάση Δεδομένων: MEDLINE
καταχωρήστε σχόλιο πρώτοι!
Πρέπει πρώτα να συνδεθείτε