Cited by DisDock: A Deep Learning Method for Metal Ion-Protein Redocking

Recent Advances and Future Challenges in Predictive Modelling of Metalloproteins by Artificial Intelligence DOI

Soohyeong Kim,

Wonseok Lee,

Hugh I. Kim

et al.

Molecules and Cells, Journal Year: 2025, Volume and Issue: unknown, P. 100191 - 100191

Published: Feb. 1, 2025

Metal coordination is essential for structural/catalytic functions of metalloproteins that mediate a wide range biological processes in living organisms. Advances bioinformatics have significantly enhanced our understanding metal-binding sites and their functional roles metalloproteins. State-of-the-art computational models developed seamlessly integrate protein sequence structural data to unravel the complexities metal environments. Our goal this mini-review give an overview these tools highlight current challenges (predicting dynamic sites, determining metalation states, designing intricate networks) remaining predictive sites. Addressing will not only deepen knowledge natural but also accelerate development artificial with novel precisely engineered functionalities.

Language: Английский

Citations

Iron: Life’s primeval transition metal DOI

Jena E. Johnson, Theodore M. Present, Joan Selverstone Valentine

et al.

Proceedings of the National Academy of Sciences, Journal Year: 2024, Volume and Issue: 121(38)

Published: Sept. 9, 2024

Modern life requires many different metal ions, which enable diverse biochemical functions. It is commonly assumed that ions’ environmental availabilities controlled the evolution of early life. We argue can only explore chemistry encounters, and fortuitous chemical interactions between ions biological compounds be selected for if they first occur sufficiently frequently. calculated maximal transition ion concentrations in ancient ocean, determining amounts biologically important were orders magnitude lower than ferrous iron. Under such conditions, primitive bioligands would predominantly interact with Fe(II). While other metals certain environments may have provided evolutionary opportunities, capacities Fe(II), Fe–S clusters, or plentiful magnesium calcium could satisfied all functions needed by Primitive organisms used Fe(II) exclusively their requirements.

Language: Английский

Citations

MetalProGNet: a structure-based deep graph model for metalloprotein–ligand interaction predictions DOI

Dejun Jiang,

Zhaofeng Ye,

Chang‐Yu Hsieh

et al.

Chemical Science, Journal Year: 2023, Volume and Issue: 14(8), P. 2054 - 2069

Published: Jan. 1, 2023

Metalloproteins play essential roles in various biological processes ranging from reaction catalysis to free radical scavenging, and they are also pertinent numerous pathologies including cancer, HIV infection,and inflammation.

Language: Английский

Citations

Metalloproteins and metalloproteomics in health and disease DOI

Iman Ibrahim

Advances in protein chemistry and structural biology, Journal Year: 2024, Volume and Issue: unknown, P. 123 - 176

Published: Jan. 1, 2024

Language: Английский

Citations

A database overview of metal-coordination distances in metalloproteins DOI

Milana Bazayeva, Claudia Andreini, Antonio Rosato

et al.

Acta Crystallographica Section D Structural Biology, Journal Year: 2024, Volume and Issue: 80(5), P. 362 - 376

Published: April 29, 2024

Metalloproteins are ubiquitous in all living organisms and take part a very wide range of biological processes. For this reason, their experimental characterization is crucial to obtain improved knowledge structure functions. The three-dimensional represents highly relevant information since it provides insight into the interaction between metal ion(s) protein fold. Such interactions determine chemical reactivity bound metal. available PDB structures can contain errors due factors such as poor resolution radiation damage. A lack use distance restraints during refinement validation process also impacts quality. Here, aim was thorough overview distribution distances ions donor atoms through statistical analysis data set based on more than 115 000 metal-binding sites proteins. This not only produced reference that be used by experimentalists support structure-determination process, for example restraints, but resulted an how coordination occurs different metals nature binding interactions. In particular, features carboxylate were inspected, which type commonly present nearly metals.

Language: Английский

Citations

DisDock: A Deep Learning Method for Metal Ion‐Protein Redocking DOI

M.-R. Lin,

Keqiao Li,

Yuan Zhang

et al.

Proteins Structure Function and Bioinformatics, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 22, 2025

ABSTRACT The structures of metalloproteins are essential for comprehending their functions and interactions. breakthrough AlphaFold has made it possible to predict protein with experimental accuracy. However, the type metal ion that a metalloprotein binds binding structure still not readily available, even predicted structure. In this study, we present DisDock, deep learning method predicting protein‐metal docking. DisDock takes distogram randomly initialized protein‐ligand configuration as input outputs complex. It combines U‐net architecture self‐attention modules enhance model performance. Taking inspiration from physical principle atoms in closer proximity display stronger mutual attraction, predictor capitalizes on geometric information uncover latent characteristics indicative atom To train our model, employ high‐quality dataset sourced Mother All Databases (MOAD). Experimental results demonstrate approach outperforms other existing methods prediction accuracy various types ions.

Language: Английский

Citations

A bioinformatics approach to the design of minimal biomimetic metal-binding peptides DOI

Mun Hon Cheah, Claudia Spallacci, Marco Chino

et al.

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: March 10, 2025

Abstract Nature-inspired or biomimetic catalyst aims to reach the high catalytic performance and selectivity of natural enzymes while possessing chemical stability processability synthetic catalysts. A promising strategy for designing catalysts holds on mimicking structure enzyme active site. This can either entail complicated total synthesis a design peptide sequences, able self-assemble in presence metal ions, thus forming metallo-peptide complexes that mimic sites enzymes. Using bioinformatics approach, we designed minimal made up eight amino acids (H4pep) act as functional trinuclear Cu site laccase enzyme. Cu(II) binding H4pep results formation Cu²⁺(H4pep)₂ complex with β-sheet secondary structure, reduce O₂. Our study demonstrates viability potential using short peptides Teaser peptide, via bioinformatics, effectively mimics copper O₂ reduction. MAIN TEXT

Language: Английский

Citations

deep-Sep: a deep learning-based method for fast and accurate prediction of selenoprotein genes in bacteria DOI

Yao Xiao, Yan Zhang

mSystems, Journal Year: 2025, Volume and Issue: unknown

Published: March 10, 2025

ABSTRACT Selenoproteins are a special group of proteins with major roles in cellular antioxidant defense. They contain the 21st amino acid selenocysteine (Sec) active sites, which is encoded by an in-frame UGA codon. Compared to eukaryotes, identification selenoprotein genes bacteria remains challenging due absence effective strategy for distinguishing Sec-encoding codon from normal stop signal. In this study, we have developed deep learning-based algorithm, deep-Sep, quickly and precisely identifying bacterial genomic sequences. This algorithm uses Transformer-based neural network architecture construct optimal model detecting codons homology search-based remove additional false positives. During training testing stages, deep-Sep has demonstrated commendable performance, including F 1 score 0.939 area under receiver operating characteristic curve 0.987. Furthermore, when applied 20 genomes as independent test data sets, exhibited remarkable capability both known new genes, significantly outperforms existing state-of-the-art method. Our proved be powerful tool comprehensively characterizing genomes, should not only assist accurate annotation genome sequencing projects but also provide insights deeper understanding selenium bacteria. IMPORTANCE Selenium essential micronutrient present selenoproteins form Sec, rare opal UGA. Identification all vital importance investigating functions nature. Previous strategies predicting mainly relied on cis -acting Sec insertion sequence (SECIS) element within mRNAs. However, complexity variability SECIS elements, recognition still challenge genomes. We predict sequences, demonstrates superior performance compared currently available methods. can utilized either web-based or local (standalone) modes, serving promising complete set

Language: Английский

Citations

Benchmarking Zinc-Binding Site Predictors: A Comparative Analysis of Structure-Based Approaches DOI

Cosimo Ciofalo,

Vincenzo Laveglia, Claudia Andreini

et al.

Journal of Chemical Information and Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: May 15, 2025

Metalloproteins play crucial physiological roles across all domains of life, relying on metal ions for structural stability and catalytic activity. In recent years, computational approaches have emerged as powerful increasingly reliable tools predicting metal-binding sites in metalloproteins, enabling their application the challenging field metalloproteomics. Given growing number available tools, it is timely to design a reproducible approach characterize performance specific usage scenarios. Thus, this study, we selected some state-of-the-art structure-based predictors zinc-binding evaluated two data sets: experimental apoprotein structures models generated by AlphaFold. Our results indicate that pose significant challenges sites. For these systems, achieved lower-than-expected due rearrangements occurring upon metalation. Conversely, predictions based AlphaFold yielded significantly better results, suggesting they more closely resemble holo forms metalloproteins. findings highlight great potential site advancing research

Language: Английский

Citations

Unlocking Precision Docking for Metalloproteins DOI

Camila M. Clemente,

Juan Miguel Monterrubio Prieto,

Marcelo A. Martí

et al.

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: 64(5), P. 1581 - 1592

Published: Feb. 19, 2024

Metalloproteins play a fundamental role in molecular biology, contributing to various biological processes. However, the discovery of high-affinity ligands targeting metalloproteins has been delayed due, part, lack suitable tools and data. Molecular docking, widely used technique for virtual screening small-molecule ligand interactions with proteins, often faces challenges when applied due particular nature metal bond. To address these limitations associated docking metalloproteins, we introduce knowledge-driven approach known as "metalloprotein bias docking" (MBD), which extends AutoDock Bias technique. We assembled comprehensive data set metalloprotein-ligand complexes from 15 different metalloprotein families, encompassing Ca, Co, Fe, Mg, Mn, Zn ions. Subsequently, conducted performance analysis our MBD method compared it conventional (CD) program AutoDock4, targets within set. Our results demonstrate that outperforms CD, significantly enhancing accuracy, selectivity, precision pose prediction. Additionally, observed positive correlation between predicted free energies corresponding experimental values. These findings underscore potential valuable tool effective exploration interactions.

Language: Английский

Citations