The success rate of processed predicted models in molecular replacement: implications for experimental phasing in the AlphaFold era.

The availability of highly accurate protein structure predictions from AlphaFold2 (AF2) and similar tools has hugely expanded the applicability of molecular replacement (MR) for crystal structure solution. Many structures can be solved routinely using raw models, structures processed to remove unreliable parts or models split into distinct structural units. There is therefore an open question around how many and which cases still require experimental phasing methods such as single-wavelength anomalous diffraction (SAD). Here, this question is addressed using a large set of PDB depositions that were solved by SAD. A large majority (87%) could be solved using unedited or minimally edited AF2 predictions. A further 18 (4%) yield straightforwardly to MR after splitting of the AF2 prediction using Slice'N'Dice, although different splitting methods succeeded on slightly different sets of cases. It is also found that further unique targets can be solved by alternative modelling approaches such as ESMFold (four cases), alternative MR approaches such as ARCIMBOLDO and AMPLE (two cases each), and multimeric model building with AlphaFold-Multimer or UniFold (three cases). Ultimately, only 12 cases, or 3% of the SAD-phased set, did not yield to any form of MR tested here, offering valuable hints as to the number and the characteristics of cases where experimental phasing remains essential for macromolecular structure solution.

Evaluation of AlphaFold3 for the fatty acids docking to human fatty acid-binding proteins.

Human fatty acid-binding proteins (FABPs) are involved in many aspects of lipid metabolism, such as the uptake, transport, and storage of lipophilic molecules, as well as cellular functions. Understanding how FABPs recognize fatty acids (FAs) is crucial for identifying FABP function and applications, such as in inhibitor design or biomarker development. The recently developed AlphaFold3 (AF3) demonstrates significantly higher accuracy than other prediction tools, particularly in predicting protein-ligand interactions with state-of-the-art docking tools. Studies on whether AF3 can be used to identify the FAs of FABP are lacking. To assess the accuracy of FA docking to FABPs using AF3, models of FA docked into FABP generated using AF3 were compared with experimentally determined FA-bound FABP structures. FA ligands in AF3 structures docked reliably into the FA-binding pocket of FABPs; however, the detailed binding configuration of most FA ligands docked into FABPs and the interaction between FA and FABP determined using AF3 and experimentally differed. These results will aid in understanding FA docking to FABPs and other FA-binding proteins using AF3.

Artificial intelligence alphafold model for molecular biology and drug discovery: a machine-learning-driven informatics investigation.

AlphaFold model has reshaped biological research. However, vast unstructured data in the entire AlphaFold field requires further analysis to fully understand the current research landscape and guide future exploration. Thus, this scientometric analysis aimed to identify critical research clusters, track emerging trends, and highlight underexplored areas in this field by utilizing machine-learning-driven informatics methods. Quantitative statistical analysis reveals that the AlphaFold field is enjoying an astonishing development trend (Annual Growth Rate = 180.13%) and global collaboration (International Co-authorship = 33.33%). Unsupervised clustering algorithm, time series tracking, and global impact assessment point out that Cluster 3 (Artificial Intelligence-Powered Advancements in AlphaFold for Structural Biology) has the greatest influence (Average Citation = 48.36 ± 184.98). Additionally, regression curve and hotspot burst analysis highlight "structure prediction" (s = 12.40, R2 = 0.9480, p = 0.0051), "artificial intelligence" (s = 5.00, R2 = 0.8096, p = 0.0375), "drug discovery" (s = 1.90, R2 = 0.7987, p = 0.0409), and "molecular dynamics" (s = 2.40, R2 = 0.8000, p = 0.0405) as core hotspots driving the research frontier. More importantly, the Walktrap algorithm further reveals that "structure prediction, artificial intelligence, molecular dynamics" (Relevance Percentage[RP] = 100%, Development Percentage[DP] = 25.0%), "sars-cov-2, covid-19, vaccine design" (RP = 97.8%, DP = 37.5%), and "homology modeling, virtual screening, membrane protein" (RP = 89.9%, DP = 26.1%) are closely intertwined with the AlphaFold model but remain underexplored, which implies a broad exploration space. In conclusion, through the machine-learning-driven informatics methods, this scientometric analysis offers an objective and comprehensive overview of global AlphaFold research, identifying critical research clusters and hotspots while prospectively pointing out underexplored critical areas.

Empowering AlphaFold2 for protein conformation selective drug discovery with AlphaFold2-RAVE.

Small-molecule drug design hinges on obtaining co-crystallized ligand-protein structures. Despite AlphaFold2's strides in protein native structure prediction, its focus on apo structures overlooks ligands and associated holo structures. Moreover, designing selective drugs often benefits from the targeting of diverse metastable conformations. Therefore, direct application of AlphaFold2 models in virtual screening and drug discovery remains tentative. Here, we demonstrate an AlphaFold2-based framework combined with all-atom enhanced sampling molecular dynamics and Induced Fit docking, named AF2RAVE-Glide, to conduct computational model-based small-molecule binding of metastable protein kinase conformations, initiated from protein sequences. We demonstrate the AF2RAVE-Glide workflow on three different mammalian protein kinases and their type I and II inhibitors, with special emphasis on binding of known type II kinase inhibitors which target the metastable classical DFG-out state. These states are not easy to sample from AlphaFold2. Here, we demonstrate how with AF2RAVE these metastable conformations can be sampled for different kinases with high enough accuracy to enable subsequent docking of known type II kinase inhibitors with more than 50% success rates across docking calculations. We believe the protocol should be deployable for other kinases and more proteins generally.

Structural prediction of GluN3 NMDA receptors.

N-methyl-D-aspartate (NMDA) receptors are heterotetrametric ion channels composed of two obligatory GluN1 subunits and two alternative GluN2 or GluN3 subunits, forming GluN1-N2, GluN1-N3, and GluN1-N2-N3 type of NMDA receptors. Extensive research has focused on the functional and structural properties of conventional GluN1-GluN2 NMDA receptors due to their early discovery and high expression levels. However, the knowledge of unconventional GluN1-N3 NMDA receptors remains limited. In this study, we modeled the GluN1-N3A, GluN1-N3B, and GluN1-N3A-N3B NMDA receptors using deep-learned protein-language predication algorithms AlphaFold and RoseTTAFold All-Atom. We then compared these structures with GluN1-N2 and GluN1-N3A receptor cryo-EM structures and found that GluN1-N3 receptors have distinct properties in subunit arrangement, domain swap, and domain interaction. Furthermore, we predicted the agonist- or antagonist-bound structures, highlighting the key molecular-residue interactions. Our findings shed new light on the structural and functional diversity of NMDA receptors and provide a new direction for drug development. This study uses advanced AI algorithms to model GluN1-N3 NMDA receptors, revealing unique structural properties and interactions compared to conventional GluN1-N2 receptors. By highlighting key molecular-residue interactions and predicting ligand-bound structures, our research enhances the understanding of NMDA receptor diversity and offers new insights for targeted drug development.

GraphNABP: Identifying nucleic acid-binding proteins with protein graphs and protein language models.

The computational identification of nucleic acid-binding proteins (NABP) is of great significance for understanding the mechanisms of these biological activities and drug discovery. Although a bunch of sequence-based methods have been proposed to predict NABP and achieved promising performance, the structure information is often overlooked. On the other hand, the power of popular protein language models (pLM) has seldom been harnessed for predicting NABPs. In this study, we propose a novel framework called GraphNABP, to predict NABP by integrating sequence and predicted 3D structure information. Specifically, sequence embeddings and protein molecular graphs were first obtained from ProtT5 protein language model and predicted 3D structures, respectively. Then, graph attention (GAT) and bidirectional long short-term memory (BiLSTM) neural networks were used to enhance feature representations. Finally, a fully connected layer is used to predict NABPs. To the best of our knowledge, this is the first time to integrate AlphaFold and protein language models for the prediction of NABPs. The performances on multiple independent test sets indicate that GraphNABP outperforms other state-of-the-art methods. Our results demonstrate the effectiveness of pLM embeddings and structural information for NABP prediction. The codes and data used in this study are available at https://github.com/lixiangli01/GraphNABP.

Modular Structure and Polymerization Status of GABAA Receptors Illustrated with EM Analysis and AlphaFold2 Prediction.

Type-A γ-aminobutyric acid (GABAA) receptors are channel proteins crucial to mediating neuronal balance in the central nervous system (CNS). The structure of GABAA receptors allows for multiple binding sites and is key to drug development. Yet the formation mechanism of the receptor's distinctive pentameric structure is still unknown. This study aims to investigate the role of three predominant subunits of the human GABAA receptor in the formation of protein pentamers. Through purifying and refolding the protein fragments of the GABAA receptor α1, ß2, and γ2 subunits, the particle structures were visualised with negative staining electron microscopy (EM). To aid the analysis, AlphaFold2 was used to compare the structures. Results show that α1 and ß2 subunit fragments successfully formed homo-oligomers, particularly homopentameric structures, while the predominant heteropentameric GABAA receptor was also replicated through the combination of the three subunits. However, homopentameric structures were not observed with the γ2 subunit proteins. A comparison of the AlphaFold2 predictions and the previously obtained cryo-EM structures presents new insights into the subunits' modular structure and polymerization status. By performing experimental and computational studies, a deeper understanding of the complex structure of GABAA receptors is provided. Hopefully, this study can pave the way to developing novel therapeutics for neuropsychiatric diseases.

Reliability of AlphaFold2 Models in Virtual Drug Screening: A Focus on Selected Class A GPCRs.

Protein three-dimensional (3D) structure prediction is one of the most challenging issues in the field of computational biochemistry, which has overwhelmed scientists for almost half a century. A significant breakthrough in structural biology has been established by developing the artificial intelligence (AI) system AlphaFold2 (AF2). The AF2 system provides a state-of-the-art prediction of protein structures from nearly all known protein sequences with high accuracy. This study examined the reliability of AF2 models compared to the experimental structures in drug discovery, focusing on one of the most common protein drug-targeted classes known as G protein-coupled receptors (GPCRs) class A. A total of 32 representative protein targets were selected, including experimental structures of X-ray crystallographic and Cryo-EM structures and their corresponding AF2 models. The quality of AF2 models was assessed using different structure validation tools, including the pLDDT score, RMSD value, MolProbity score, percentage of Ramachandran favored, QMEAN Z-score, and QMEANDisCo Global. The molecular docking was performed using the Genetic Optimization for Ligand Docking (GOLD) software. The AF2 models' reliability in virtual drug screening was determined by their ability to predict the ligand binding poses closest to the native binding pose by assessing the Root Mean Square Deviation (RMSD) metric and docking scoring function. The quality of the docking and scoring function was evaluated using the enrichment factor (EF). Furthermore, the capability of using AF2 models in molecular docking to identify hits with key protein-ligand interactions was analyzed. The posing power results showed that the AF2 models successfully predicted ligand binding poses (RMSD < 2 Å). However, they exhibited lower screening power, with average EF values of 2.24, 2.42, and 1.82 for X-ray, Cryo-EM, and AF2 structures, respectively. Moreover, our study revealed that molecular docking using AF2 models can identify competitive inhibitors. In conclusion, this study found that AF2 models provided docking results comparable to experimental structures, particularly for certain GPCR targets, and could potentially significantly impact drug discovery.

A deep learning framework combining molecular image and protein structural representations identifies candidate drugs for pain.

Artificial intelligence (AI) and deep learning technologies hold promise for identifying effective drugs for human diseases, including pain. Here, we present an interpretable deep-learning-based ligand image- and receptor's three-dimensional (3D)-structure-aware framework to predict compound-protein interactions (LISA-CPI). LISA-CPI integrates an unsupervised deep-learning-based molecular image representation (ImageMol) of ligands and an advanced AlphaFold2-based algorithm (Evoformer). We demonstrated that LISA-CPI achieved ∼20% improvement in the average mean absolute error (MAE) compared to state-of-the-art models on experimental CPIs connecting 104,969 ligands and 33 G-protein-coupled receptors (GPCRs). Using LISA-CPI, we prioritized potential repurposable drugs (e.g., methylergometrine) and identified candidate gut-microbiota-derived metabolites (e.g., citicoline) for potential treatment of pain via specifically targeting human GPCRs. In summary, we presented that the integration of molecular image and protein 3D structural representations using a deep learning framework offers a powerful computational drug discovery tool for treating pain and other complex diseases if broadly applied.

Characterization of lignin-degrading enzyme PmdC, which catalyzes a key step in the synthesis of polymer precursor 2-pyrone-4,6-dicarboxylic acid.

Pyrone-2,4-dicarboxylic acid (PDC) is a valuable polymer precursor that can be derived from the microbial degradation of lignin. The key enzyme in the microbial production of PDC is 4-carboxy-2-hydroxymuconate-6-semialdehyde (CHMS) dehydrogenase, which acts on the substrate CHMS. We present the crystal structure of CHMS dehydrogenase (PmdC from Comamonas testosteroni) bound to the cofactor NADP, shedding light on its three-dimensional architecture, and revealing residues responsible for binding NADP. Using a combination of structural homology, molecular docking, and quantum chemistry calculations, we have predicted the binding site of CHMS. Key histidine residues in a conserved sequence are identified as crucial for binding the hydroxyl group of CHMS and facilitating dehydrogenation with NADP. Mutating these histidine residues results in a loss of enzyme activity, leading to a proposed model for the enzyme's mechanism. These findings are expected to help guide efforts in protein and metabolic engineering to enhance PDC yields in biological routes to polymer feedstock synthesis.

Protein expression, purification, crystallization and crystallographic studies of BPSL0741 from Burkholderia pseudomallei.

Burkholderia pseudomallei is the causative agent of the lethal disease melioidosis. This bacterium infects animals and humans and is increasingly resistant to multiple antibiotics. Recently, genes associated with survival of the bacterium in the infected host have been identified. One of these genes, bpsl0741, is annotated as a hypothetical protein of 185 amino acids. Here, recombinant BPSL0741 (rBPSL0741) protein was expressed, purified, verified by mass spectrometry, crystallized and analyzed by X-ray diffraction. rBPSL0741 was crystallized by vapor diffusion using a reservoir solution consisting of 0.2 M ammonium acetate, 0.1 M sodium acetate trihydrate pH 4.6, 30% PEG 4000. The crystals diffracted to 2.1 Šresolution using an in-house X-ray diffractometer and belonged to an orthorhombic space group, with unit-cell parameters a = 62.92, b = 64.57, c = 89.16 Å. The Matthews coefficient (VM) was calculated to be 2.18 Å3 Da-1, suggesting the presence of two molecules per asymmetric unit and an estimated solvent content of 43.5%. The crystal was deemed to be suitable for further structural studies, which are currently ongoing.

Crystal structure of guanosine 5'-monophosphate synthetase from the thermophilic bacterium Thermus thermophilus HB8.

Guanosine 5'-monophosphate (GMP) synthetase (GuaA) catalyzes the last step of GMP synthesis in the purine nucleotide biosynthetic pathway. This enzyme catalyzes a reaction in which xanthine 5'-monophosphate (XMP) is converted to GMP in the presence of Gln and ATP through an adenyl-XMP intermediate. A structure of an XMP-bound form of GuaA from the domain Bacteria has not yet been determined. In this study, the crystal structure of an XMP-bound form of GuaA from the thermophilic bacterium Thermus thermophilus HB8 (TtGuaA) was determined at a resolution of 2.20 Šand that of an apo form of TtGuaA was determined at 2.10 Šresolution. TtGuaA forms a homodimer, and the monomer is composed of three domains, which is a typical structure for GuaA. Disordered regions in the crystal structure were obtained from the AlphaFold2-predicted model structure, and a model with substrates (Gln, XMP and ATP) was constructed for molecular-dynamics (MD) simulations. The structural fluctuations of the TtGuaA dimer as well as the interactions between the active-site residues were analyzed by MD simulations.

PPI-hotspotID for detecting protein-protein interaction hot spots from the free protein structure.

Experimental detection of residues critical for protein-protein interactions (PPI) is a time-consuming, costly, and labor-intensive process. Hence, high-throughput PPI-hot spot prediction methods have been developed, but they have been validated using relatively small datasets, which may compromise their predictive reliability. Here, we introduce PPI-hotspotID, a novel method for identifying PPI-hot spots using the free protein structure, and validated it on the largest collection of experimentally confirmed PPI-hot spots to date. We explored the possibility of detecting PPI-hot spots using (i) FTMap in the PPI mode, which identifies hot spots on protein-protein interfaces from the free protein structure, and (ii) the interface residues predicted by AlphaFold-Multimer. PPI-hotspotID yielded better performance than FTMap and SPOTONE, a webserver for predicting PPI-hot spots given the protein sequence. When combined with the AlphaFold-Multimer-predicted interface residues, PPI-hotspotID yielded better performance than either method alone. Furthermore, we experimentally verified several PPI-hotspotID-predicted PPI-hot spots of eukaryotic elongation factor 2. Notably, PPI-hotspotID can reveal PPI-hot spots not obvious from complex structures, including those in indirect contact with binding partners. PPI-hotspotID serves as a valuable tool for understanding PPI mechanisms and aiding drug design. It is available as a web server (https://ppihotspotid.limlab.dnsalias.org/) and open-source code (https://github.com/wrigjz/ppihotspotid/).

AI-based IsAb2.0 for antibody design.

Therapeutic antibody design has garnered widespread attention, highlighting its interdisciplinary importance. Advancements in technology emphasize the critical role of designing nanobodies and humanized antibodies in antibody engineering. However, current experimental methods are costly and time-consuming. Computational approaches, while progressing, faced limitations due to insufficient structural data and the absence of a standardized protocol. To tackle these challenges, our lab previously developed IsAb1.0, an in silico antibody design protocol. Yet, IsAb1.0 lacked accuracy, had a complex procedure, and required extensive antibody bioinformation. Moreover, it overlooked nanobody and humanized antibody design, hindering therapeutic antibody development. Building upon IsAb1.0, we enhanced our design protocol with artificial intelligence methods to create IsAb2.0. IsAb2.0 utilized AlphaFold-Multimer (2.3/3.0) for accurate modeling and complex construction without templates and employed the precise FlexddG method for in silico antibody optimization. Validated through optimization of a humanized nanobody J3 (HuJ3) targeting HIV-1 gp120, IsAb2.0 predicted five mutations that can improve HuJ3-gp120 binding affinity. These predictions were confirmed by commercial software and validated through binding and neutralization assays. IsAb2.0 streamlined antibody design, offering insights into future techniques to accelerate immunotherapy development.

AlzDiscovery: A computational tool to identify Alzheimer's disease-causing missense mutations using protein structure information.

Alzheimer's disease (AD) is one of the most common forms of dementia and neurodegenerative diseases, characterized by the formation of neuritic plaques and neurofibrillary tangles. Many different proteins participate in this complicated pathogenic mechanism, and missense mutations can alter the folding and functions of these proteins, significantly increasing the risk of AD. However, many methods to identify AD-causing variants did not consider the effect of mutations from the perspective of a protein three-dimensional environment. Here, we present a machine learning-based analysis to classify the AD-causing mutations from their benign counterparts in 21 AD-related proteins leveraging both sequence- and structure-based features. Using computational tools to estimate the effect of mutations on protein stability, we first observed a bias of the pathogenic mutations with significant destabilizing effects on family AD-related proteins. Combining this insight, we built a generic predictive model, and improved the performance by tuning the sample weights in the training process. Our final model achieved the performance on area under the receiver operating characteristic curve up to 0.95 in the blind test and 0.70 in an independent clinical validation, outperforming all the state-of-the-art methods. Feature interpretation indicated that the hydrophobic environment and polar interaction contacts were crucial to the decision on pathogenic phenotypes of missense mutations. Finally, we presented a user-friendly web server, AlzDiscovery, for researchers to browse the predicted phenotypes of all possible missense mutations on these 21 AD-related proteins. Our study will be a valuable resource for AD screening and the development of personalized treatment.

Alternating access of a bacterial homolog of neurotransmitter: sodium symporters determined from AlphaFold2 ensembles and DEER spectroscopy.

Neurotransmitter:sodium symporters (NSSs) play critical roles in neural signaling by regulating neurotransmitter uptake into cells powered by sodium electrochemical gradients. Bacterial NSSs orthologs, including MhsT from Bacillus halodurans, have emerged as model systems to understand the structural motifs of alternating access in NSSs and the extent of conservation of these motifs across the family. Here, we apply a computational/experimental methodology to illuminate the conformational landscape of MhsT alternating access. Capitalizing on our recently developed method, Sampling Protein Ensembles and Conformational Heterogeneity with AlphaFold2 (SPEACH_AF), we derived clusters of MhsT models spanning the transition from inward-facing to outward-facing conformations. Systematic application of double electron-electron resonance (DEER) spectroscopy revealed ligand-dependent movements of multiple structural motifs that underpin MhsT's conformational cycle. Remarkably, comparative DEER analysis in detergent micelles and lipid nanodiscs highlights the profound effect of the environment on the energetics of conformational changes. Through experimentally derived selection of collective variables, we present a model of ion and substrate-powered transport by MhsT consistent with the conformational cycle derived from DEER. Our findings not only advance the understanding of MhsT's function but also uncover motifs of conformational dynamics conserved within the broader context of the NSS family and within the LeuT-fold class of transporters. Importantly, our methodological blueprint introduces an approach that can be applied across a diverse spectrum of transporters to describe their conformational landscapes.

An outlook on structural biology after AlphaFold: tools, limits and perspectives.

AlphaFold and similar groundbreaking, AI-based tools, have revolutionized the field of structural bioinformatics, with their remarkable accuracy in ab-initio protein structure prediction. This success has catalyzed the development of new software and pipelines aimed at incorporating AlphaFold's predictions, often focusing on addressing the algorithm's remaining challenges. Here, we present the current landscape of structural bioinformatics shaped by AlphaFold, and discuss how the field is dynamically responding to this revolution, with new software, methods, and pipelines. While the excitement around AI-based tools led to their widespread application, it is essential to acknowledge that their practical success hinges on their integration into established protocols within structural bioinformatics, often neglected in the context of AI-driven advancements. Indeed, user-driven intervention is still as pivotal in the structure prediction process as in complementing state-of-the-art algorithms with functional and biological knowledge.

Mass spectrometry-complemented molecular modeling predicts the interaction interface for a camelid single-domain antibody targeting the Plasmodium falciparum circumsporozoite protein's C-terminal domain.

Background: Bioanalytical methods that enable rapid and high-detail characterization of binding specificities and strengths of protein complexes with low sample consumption are highly desired. The interaction between a camelid single domain antibody (sdAbCSP1) and its target antigen (PfCSP-Cext) was selected as a model system to provide proof-of-principle for the here described methodology. Research design and methods: The structure of the sdAbCSP1 - PfCSP-Cext complex was modeled using AlphaFold2. The recombinantly expressed proteins, sdAbCSP1, PfCSP-Cext, and the sdAbCSP1 - PfCSP-Cext complex, were subjected to limited proteolysis and mass spectrometric peptide analysis. ITEM MS (Intact Transition Epitope Mapping Mass Spectrometry) and ITC (Isothermal Titration Calorimetry) were applied to determine stoichiometry and binding strength. Results: The paratope of sdAbCSP1 mainly consists of its CDR3 (aa100-118). PfCSP-Cext's epitope is assembled from its α-helix (aa40-52) and opposing loop (aa83-90). PfCSP-Cext's GluC cleavage sites E46 and E58 were shielded by complex formation, confirming the predicted epitope. Likewise, sdAbCSP1's tryptic cleavage sites R105 and R108 were shielded by complex formation, confirming the predicted paratope. ITEM MS determined the 1:1 stoichiometry and the high complex binding strength, exemplified by the gas phase dissociation reaction enthalpy of 50.2 kJ/mol. The in-solution complex dissociation constant is 5 × 10-10 M. Conclusions: Combining AlphaFold2 modeling with mass spectrometry/limited proteolysis generated a trustworthy model for the sdAbCSP1 - PfCSP-Cext complex interaction interface.

Novel structural insights at the extracellular plant-pathogen interface.

Plant pathogens represent a critical threat to global agriculture and food security, particularly under the pressures of climate change and reduced agrochemical use. Most plant pathogens initially colonize the extracellular space or apoplast and understanding the host-pathogen interactions that occur here is vital for engineering sustainable disease resistance in crops. Structural biology has played important roles in elucidating molecular mechanisms underpinning plant-pathogen interactions but only few studies have reported structures of extracellular complexes. This article highlights these resolved extracellular complexes by describing the insights gained from the solved structures of complexes consisting of CERK1-chitin, FLS2-flg22-BAK1, RXEG1-XEG1-BAK1 and PGIP2-FpPG. Finally, we discuss the potential of AI-based structure prediction platforms like AlphaFold as an alternative hypothesis generator to rapidly advance our molecular understanding of plant pathology and develop novel strategies to increase crop resilience against disease.

An AI-informed NMR structure reveals an extraordinary LETM1 F-EF-hand domain that functions as a two-way regulator of mitochondrial calcium.

AlphaFold can accurately predict static protein structures but does not account for solvent conditions. Human leucine zipper EF-hand transmembrane protein-1 (LETM1) has one sequence-identifiable EF-hand but how calcium (Ca2+) affects structure and function remains enigmatic. Here, we used highly confident AlphaFold Cα predictions to guide nuclear Overhauser effect (NOE) assignments and structure calculation of the LETM1 EF-hand in the presence of Ca2+. The resultant NMR structure exposes pairing between a partial loop-helix and full helix-loop-helix, forming an unprecedented F-EF-hand with non-canonical Ca2+ coordination but enhanced hydrophobicity for protein interactions compared to calmodulin. The structure also reveals the basis for pH sensing at the link between canonical and partial EF-hands. Functionally, mutations that augmented or weakened Ca2+ binding increased or decreased matrix Ca2+, respectively, establishing F-EF as a two-way mitochondrial Ca2+ regulator. Thus, we show how to synergize AI prediction with NMR data, elucidating a solution-specific and extraordinary LETM1 F-EF-hand.