SEARCH · PubMed Health

Results for “AlphaFold3”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

13 recordsLinked to original sources

PubMed2026

Stable simulations do not guarantee functional engagement: a case study of off-target prediction for Seladelpar and Zanamivir.

Identifying off-target interactions of approved drugs is important to anticipate side effects and uncover repurposing opportunities. Computational pipelines combining structural homology, structure prediction, and molecular dynamics (MD) simulations offer a promising strategy, but it remains unclear whether stable, control-like MD trajectories reliably indicate functional engagement. We examined this in a case study of two approved drugs. Using the Evolutionary Classification of Protein Domains (ECOD) framework to select candidate off-targets, we modeled each drug-protein complex as two independent AlphaFold3 models and simulated both by MD, for Seladelpar (a PPARδ agonist) and Zanamivir, an influenza neuraminidase inhibitor that also inhibits human Sialidase-2 (NEU2). Candidates were ranked by the similarity of global MD descriptors to the on-target control. For Seladelpar, the three top-ranked candidates (FXR, RARγ, ERRγ) were tested experimentally; the Zanamivir set was analyzed computationally only. None showed measurable activity in reporter or thermal shift assays, despite stable simulations and descriptor values comparable to the control. Including PPARα and PPARγ as weak-positive comparators, these descriptors did not rank genuine interactions closer to the control than inactive candidates. Residue-level comparison with experimental structures showed the predicted poses reproduced only part of the canonical contacts. Where experimental drug-bound structures existed, AlphaFold3 reproduced the pose for PPARα but not PPARγ, and its per-model confidence did not track pose accuracy. Within this case study, the specific global descriptors examined reflect complex stability rather than functional engagement, which does not mean MD-based approaches cannot make this distinction.

Zanamivir↗

PubMed2026

Comprehensive evaluation of AlphaFold/OpenFold prediction of experimentally unresolved proteins through novel metrics.

Predicting accurate protein structures is essential for understanding molecular mechanisms, interpreting the impact of sequence variation, and supporting translational applications ranging from drug discovery to clinical genomics. Recent advances in deep-learning-based predictors such as AlphaFold2, OpenFold, and AlphaFold3 have transformed structural biology, enabling routine in silico modeling even for challenging or previously uncharacterized proteins. However, systematic benchmarking of these tools-especially for novel targets and single amino acid variants-remains limited. Conventional global metrics often fail to capture biologically meaningful discrepancies. By evaluating multiple implementations of AlphaFold2 and OpenFold, together with ColabFold and the AlphaFold3 server, across 10 different proteins and 222 single amino acid protein variants encompassing a wide range of sizes, structures, and functions, we show that although widely used global indicators-like mean pLDDT, pTM-score, and RMSD-frequently suggest comparable performance, substantial local-level differences remain elusive. To address this gap, we introduce a comparative framework leveraging Bland-Altman agreement analysis, to evaluate per-residue Cα-confidence differences and Per-Residue profiles (PRPs), complemented by Uniform Manifold Approximation and Projection (UMAP). This approach reveals marked localized divergences, particularly within flexible or intrinsically disordered regions, where both predictor choice and single-residue substitutions trigger the largest conformational shifts. We further demonstrate that using reduced homology databases has minimal impact on predicted structural quality, offering computationally efficient alternatives. Collectively, our findings underscore the importance of integrating global and residue-specific evaluations to more accurately assess robustness, agreement, and practical usability across contemporary protein structure prediction methods.

Proteins↗

PubMed2024

Structure and Methyl-lysine Binding Selectivity of the HUSH Complex Subunit MPP8.

The Human Silencing Hub (HUSH) guards the genome from the pathogenic effects of retroelement expression. Composed of MPP8, TASOR, and Periphilin-1, HUSH recognizes actively transcribed retrotransposed sequences by the presence of long (>1.5-kb) nascent transcripts without introns. HUSH recruits effectors that alter chromatin structure, degrade transcripts, and deposit transcriptionally repressive epigenetic marks. Here, we report the crystal structure of the C-terminal domain (CTD) of MPP8 necessary for HUSH activity. The MPP8 CTD consists of five ankyrin repeats followed by a domain with structural homology to the PINIT domains of Siz/PIAS-family SUMO E3 ligases. AlphaFold3 modeling of the MPP8-TASOR complex predicts that a SPOC domain and a domain with a novel fold in TASOR form extended interaction interfaces with the MPP8 CTD. Point mutations at these interfaces resulted in loss of HUSH-dependent transcriptional repression in a cell-based reporter assay, validating the AlphaFold3 model. The MPP8 chromodomain, known to bind the repressive mark H3K9me3, bound with similar or higher affinity to sequences in the H3K9 methyltransferase subunits SETDB1, ATF7IP, G9a, and GLP. Hence, MPP8 promotes heterochromatinization by recruiting H3K9 methyltransferases. Our work identifies novel structural elements in MPP8 required for HUSH complex assembly and silencing, thereby fulfilling vital functions in controlling retrotransposons.

Humans↗

PubMed2026

Structural Basis for Stepwise Substrate Transport and Disease Phenotypic Heterogeneity of the Mitochondrial ADP/ATP Carrier.

The mitochondrial ADP/ATP carrier (AAC) is essential for cellular energy metabolism and responsible for exchanging ADP for ATP across the inner mitochondrial membrane. However, the precise molecular determinants of substrate binding and the mechanisms underlying the phenotypic heterogeneity of AAC-related diseases remain poorly understood. Here, we combined AlphaFold3 predictions, molecular dynamics simulations, and experimental validation to identify and characterize a previously unrecognized ADP-binding site in AAC (site S2), which is distinct from the canonical bottom site (site S1). AlphaFold3 predictions on AAC variants with disrupted site S1 consistently placed ADP at site S2, interacting with residues R188, K92, and K96, a finding that was independently corroborated by our prior MD simulations. Systematic mutagenesis and functional analysis revealed distinct roles for site S2 residues: R188 serves as the primary phosphate-specific anchor; K92 and K96 facilitate initial recruitment and stabilization; and the aromatic ladder (Y187/Y191/F192/Y195) assists the conformational transition of ADP from anti to syn, a critical step enabling downward translocation. Functional characterization demonstrated that mutations in site S2 significantly impaired ADP transport and oxidative phosphorylation. Notably, the spatial distribution of AAC disease mutations correlates with this bipartite architecture: mild PEOA2-associated mutations cluster near site S2 and perturb the local conformation without abolishing binding, whereas lethal mutations cluster near site S1 and disrupt both ADP binding and structural integrity of the m-gate. Our findings provide new mechanistic insights into stepwise substrate transport and potential therapeutic targets for AAC-related diseases.

Mitochondrial ADP, ATP Translocases↗

PubMed2025

KSHVbook: An Information-Sharing Database for Kaposi's Sarcoma-Associated Herpesvirus.

Kaposi's sarcoma-associated herpesvirus (KSHV) is a double-stranded DNA virus belonging to the γ-herpesvirus subfamily. KSHV is the causative agent of Kaposi's sarcoma (KS), primary effusion lymphoma (PEL), multicentric Castleman's disease (MCD), and KSHV inflammatory cytokine syndrome (KICS). Since its discovery, research on KSHV has rapidly progressed, but existing information platforms relatively lack comprehensiveness and do not provide efficient analysis tools tailored for KSHV. To further promote the research on KSHV more effectively, we have developed KSHVbook (http://www.kshvbook.com), a specialized information-sharing database dedicated to KSHV. This platform offers extensive information on genes, coding sequences, proteins, and the gene regulatory region. Besides, the KSHVbook includes about 35 010 transcription factor binding sites (TFBSs), 342 010 pairs of KSHV miRNA-host target gene relationships, protein structures predicted by AlphaFold3, qPCR primers, and so on. We also develop analytical tools for viral genome regions, TFBSs, and KSHV miRNA target genes to discover previously unknown biological functions of KSHV. These analytical tools can effectively identify the potential regulatory relationships between host transcription factors and viral genes. Overall, this platform provides a centralized data resource for KSHV research by integrating multiple databases, offering accessible analysis tools, and simplifying data acquisition. The KSHVbook will continue to be updated, and more features can be found on the website.

Herpesvirus 8, Human↗

PubMed2026

Genomic and Molecular Interaction Analysis of NodD1 in a Novel Bradyrhizobium yuanmingense sp. B64 Isolate for Nodulation and Symbiosis of Legume Plants.

Rhizobial bacteria are known for their ability to fix nitrogen for leguminous plants and their essential function for sustainable agriculture. This study characterizes the taxonomic status and functional potential of the Bradyrhizobium B64 isolate using integrated genomic and molecular approaches. The whole genome of the B64 isolate was sequenced via Illumina paired-end technology. Species delimitation was performed using average nucleotide identity (ANI) and digital DNA-DNA Hybridization (dDDH). The NodD1 protein structure was modeled using AlphaFold3 and validated by Ramachandran plot analysis. Molecular docking was then conducted to evaluate interactions between NodD1 and four signaling flavonoids: Apigenin, Daidzein, Genistein, and Naringenin. Genomic analysis revealed a maximum ANI of 94.4% and dDDH values between 51.4 and 62.4%. Since these values fall below the standard prokaryotic thresholds (ANI < 95%; dDDH < 70%), the B64 isolate is identified as a novel species. Physiological assays confirmed nitrogen fixation (1.97 ppm), IAA production (3.67 ppm), and phosphate solubilization (26.10 ppm). Structural validation showed 100% of NodD1 residues in allowed regions, ensuring high model reliability. Docking simulations demonstrated strong binding affinities across all flavonoids, with binding free energies ranging from - 8.8 to - 9.0 kcal/mol. Daidzein exhibited the highest thermodynamic stability (- 9.0 kcal/mol), whereas apigenin showed the most extensive residue interaction network. The B64 isolate is a novel Bradyrhizobium species with a high symbiotic capacity. The stable NodD1-flavonoid interactions provide a molecular basis for efficient nodulation, positioning B64 as a promising candidate for developing lipo-chitooligosaccharide (LCO)-based biofertilizers.

Bradyrhizobium↗

PubMed2026

Affinity-matured B cell responses neutralizing type-I interferons underlie severe viral infections.

Autoantibodies neutralizing type-I interferons (AAN-I-IFNs) emerge as global, common, and strong determinants of a growing number of severe viral diseases. We report that AAN-I-IFNs+ patients with life-threatening COVID-19 pneumonia harbor circulating type-I IFN-specific B cells indistinguishable from patients bearing T cell tolerance defects of genetic origin. This autoimmune response mobilizes a highly diverse and stable circulating B cell response that is detected prior to severe viral infection and acquires high affinity and neutralization potential to type-I IFNs through extended somatic hypermutation. X-ray crystallography and AlphaFold3 structural analysis of hundreds of patient-derived monoclonal antibodies reveals the extended breadth of this response, targeting three major B cell epitopes covering all facets of type-I IFNs. These findings support a model in which a germinal-center-derived memory B cell response directed against type-I IFNs is established before severe viral infection, providing a core mechanism linking T cell tolerance defect to pathogenic AAN-I-IFNs underlying severe viral diseases.

Humans↗

PubMed2026

Disruption of GxxxG motifs in pATOM36 impairs biogenesis of the mitochondrial protein translocase of the outer membrane in Trypanosoma brucei.

Mitochondrial biogenesis requires efficient import of cytosolically produced proteins and correct segregation of the mitochondrial genome during cytokinesis. In Trypanosoma brucei, a parasitic protozoan with a single mitochondrion harboring a single-unit mitochondrial genome, protein import across the outer membrane is mediated by the ATOM complex. An important, yet poorly understood role is played by the integral membrane protein pATOM36 of the outer mitochondrial membrane, which is essential for both ATOM complex assembly and mitochondrial DNA segregation. Here, we combined in vivo functional mutational analysis and structural modeling to investigate the function of pATOM36. AlphaFold3-based models predict five highly tilted helices forming a funnel-shaped cavity open toward the cytoplasm, reminiscent of membrane protein insertases. In the model, the protein is sealed towards the mitochondrial intermembrane space by tight helix packing, with conserved GxxxG motifs potentially facilitating these helix-helix interactions. Progressive replacement of these glycines by isoleucines does not affect protein production or correct localization but leads to defective ATOM complex biogenesis and arrest of growth, while mitochondrial DNA segregation is largely unaffected. Based on the predicted structure, these effects can be rationalized by hydrophobic bulking that interferes with associated electrostatic interactions. This hypothesis is supported by experimental mutational analysis of the respective electrostatic interactions in the presence of native GxxxG motifs. Together, our data support the hypothesis that pATOM36 functions as an outer mitochondrial insertase and arose by convergent evolution. The GxxxG motifs, also found in unrelated yeast and human outer membrane insertases, are crucial for protein activity.

Trypanosoma brucei brucei↗

PubMed2026

Molecular characterization and biological characteristics of a highly pathogenic recombinant ALV-J strain (HUE2023) with cross-clade gp85 recombination.

Avian leukosis virus subgroup J (ALV-J) has undergone extensive diversification into phylogenetically distinct clades, yet whether recombination between these clades within the gp85 envelope glycoprotein generates variants with altered pathogenicity has received little direct investigation. A field strain (HUE2023) was recovered from breeding roosters displaying vascular tumors. The viral genome was sequenced and subjected to phylogenetic and recombination analyses. The three-dimensional structure of gp85 was predicted with AlphaFold3; electrostatic surface potentials and surface hydrophobicity were computed using the Adaptive Poisson-Boltzmann Solver and the Eisenberg hydrophobicity scale, respectively. Pathogenicity and immunosuppressive effects were assessed in Hy-Line Brown chickens. Recombination analysis revealed that HUE2023 is an inter-clade recombinant derived from Clade 1.1 (major parent: JS14NT01) and Clade 1.2 (minor parent: JS09GY3). A single-residue deletion at position 61 within receptor-binding domain 1 (RBD-1), unique to the recombinant, induced a localized conformational rearrangement that generated a concentrated electronegative surface patch and a contiguous hydrophobic pocket not observed in either parental gp85. Animal challenge showed that HUE2023 is highly pathogenic: female chickens in the high-dose group reached only 61% survival and displayed significant growth retardation (P < 0.05) together with marked immunosuppression. The recombination in the RBD-1 led to local conformational rearrangement, resulting in a concentrated and negatively charged surface area as well as a continuous hydrophobic pocket, which were never present in any of the parental gp85 sequences. These results indicate that gp85 recombination across clades can yield variants with fundamentally altered receptor-binding surfaces and argue for integrating structural surveillance into ALV-J monitoring programmes.

Animals↗

PubMed2026

Biosynthesis of Crinipellin Diterpenes in Mushroom Marasmius fiardii PR-910.

Crinipellins are a distinctive family of 5/5/5/5 tetracyclic diterpenoids previously reported exclusively from mushrooms of the genus Crinipellis. Despite extensive synthetic studies, the biosynthetic machinery responsible for crinipellin formation has remained elusive. Here, we identify the crinipellin biosynthetic gene cluster (mfd) from the mushroom Marasmius fiardii PR-910, a member of the family Marasmiaceae to which Crinipellis also belongs, although M. fiardii PR-910 itself has not been previously reported to produce crinipellins. Using a combination of site-directed mutagenesis guided by an AlphaFold3-generated structural model, stable isotope-labeling studies, density functional theory (DFT) calculations, and ab initio molecular dynamics (AIMD) simulations, the cyclization mechanism of the diterpene synthase MfdB, which constructs the fused tetraquinane scaffolds 1 and 2, was elucidated. Mutagenesis of MfdB uncovered cryptic cyclization pathways that generate structurally diverse diterpenes, including unprecedented bridged and rearranged diterpene skeletons (4-6), whose formation is supported by computational analyses, and further revealed an unusual arginine-rich diphosphate-binding architecture. Heterologous expression studies in Aspergillus oryzae and Saccharomyces cerevisiae established the oxidative functions of the cytochrome P450 enzymes MfdC, MfdD, and MfdE, leading to the production of 19 previously undescribed oxidized metabolites (16-34). Notably, MfdE, a member of the largely unexplored CYP_FUM15-like subfamily, catalyzes an unusual oxidative demethylation through C-C bond cleavage, expanding the known catalytic repertoire of fungal cytochrome P450 enzymes. Collectively, this work establishes the biosynthetic logic of crinipellin formation, reveals how terpene synthase plasticity generates cryptic diterpene scaffolds, and demonstrates how oxidative tailoring by multiple cytochrome P450 enzymes drives diterpene scaffold diversification.

Diterpenes↗

PubMed2026

A token-pruning framework enables efficient representation of the human genome for RNA modification analysis.

MOTIVATION: Modelling long genomic sequences remains challenging due to extreme sequence length, high redundancy, and the need for biological interpretability. Although Transformer-based architectures have achieved strong performance across genomic tasks, their high computational cost and reliance on fixed tokenization strategies limit their scalability and ability to focus on biologically informative regions. RESULTS: We propose ATSFormer, a token-pruning Transformer framework for efficient and biologically informed genomic sequence modelling. ATSFormer incorporates an attention-guided and parameter-free Adaptive Token Sampling (ATS) module into Transformer layers. Guided by attention-derived importance scores, ATS dynamically retains informative tokens while probabilistically discarding redundant ones, thereby reducing sequence length, FLOPs, and memory usage without introducing additional learnable parameters or extra training procedures. Importantly, the retained tokens correspond to key contributors to model predictions, enabling ATSFormer to highlight biologically meaningful sites and sequence motifs. We evaluated ATSFormer on four benchmark RNA modification datasets derived from RMVar 2.0, covering A-to-I, m1A, m5C, and m7G. Experimental results show that ATSFormer consistently outperforms existing state-of-the-art methods while achieving substantial computational savings. Furthermore, structural analysis using AlphaFold3 supports the biological relevance of the motifs identified by ATSFormer. AVAILABILITY AND IMPLEMENTATION: The source data and code are freely available at GitHub (https://github.com/1gao2/ATSFormer) and Zenodo (https://doi.org/10.5281/zenodo.21813541).

Humans↗

PubMed2025

Modeling Alternative Conformational States in CASP16.

The CASP16 Ensemble Prediction experiment assessed advances in methods for modeling proteins, nucleic acids, and their complexes in multiple conformational states. Targets included systems with experimental structures determined in two or three states, evaluated by direct comparison to experimental coordinates, as well as domain-linker-domain (D-L-D) targets assessed against statistical models from NMR and SAXS data. This paper focuses on the former class of multi-state targets. Ten ensembles were released as community challenges, including ligand-induced conformational changes, protein-DNA complexes, a trimeric protein, a stem-loop RNA, and multiple oligomeric states of a single RNA. For five targets, some groups produced reasonably accurate models of both reference states (best TM-score >0.75). However, with the exception of one protein-ligand complex (T1214), where an apo structure was available as a template, predictors generally failed to capture key structural details distinguishing the states. Overall, accuracy was significantly lower than for single-state targets in other CASP experiments. The most successful approaches generated multiple AlphaFold2 models using enhanced multiple sequence alignments and sampling protocols, followed by model quality based selection. While the AlphaFold3 server performed well on several targets, individual groups outperformed it in specific cases. By contrast, predictions for one protein-DNA complex, three RNA targets, and multiple oligomeric RNA states consistently fell short (TM-score <0.75). These results highlight both progress and persistent challenges in multi-state prediction. Despite recent advances, accurate modeling of conformational ensembles, particularly RNA and large multimeric assemblies, remains a critical frontier for structural biology.

AlphaFold2↗

PubMed2026

PGM1 deficiency is linked to sarcomeric and mitochondrial dysfunction in patient-derived iPSC-cardiomyocytes.

BACKGROUND: PGM1-congenital disorder of glycosylation (PGM1-CDG) is frequently associated with cardiomyopathy. Although galactose therapy corrects glycosylation defects, cardiac dysfunction typically persists, suggesting a glycosylation-independent mechanism. Recent evidence of mitochondrial abnormalities in PGM1-deficient human and murine heart, together with the association of PGM1 with the Z-disk protein LDB3 (ZASP/Cypher), suggests a critical role for PGM1 in cardiomyocyte structural and energetic homeostasis. We hypothesized that PGM1-related cardiomyopathy arises from a glycosylation-independent disruption of Z-disk-mitochondrial coupling driven by loss of PGM1-LDB3 interactions, resulting in mitochondrial energy failure and impaired contractile function. METHODS: Induced pluripotent stem cell-derived cardiomyocytes (iCMs) were generated from PGM1-deficient patient fibroblasts. Multielectrode array (MEA) recordings, untargeted (glyco)proteomics, and pathway analysis were performed to assess functional and molecular changes. Key findings were validated using tracer metabolomics and mitochondrial respiration assays. RESULTS: PGM1-deficient iCMs exhibited reduced beating frequency, impaired contractility, and prolonged contraction kinetics. Proteomic analyses revealed depletion of Z-disk components, including LDB3. AlphaFold3 structural modeling predicted a direct interaction between PGM1 and LDB3, implicating PGM1 in Z-disk integrity, which was confirmed in vitro. In addition, mitochondrial proteins were severely depleted, prompting us to investigate mitochondrial function. Functional validation confirmed extensive metabolic rewiring, energy depletion, and severely impaired mitochondrial respiration. Finally, the in silico drug repurposing identified possible therapeutic options that could target PGM1-deficient cardiomyopathy. CONCLUSION: Our data suggests PGM1 is key regulator of cardiomyocyte function, linking sarcomeric Z-disk integrity with mitochondrial metabolism. These mechanistic insights offer a foundation for developing targeted therapies for PGM1-CDG and potentially other cardiomyopathies involving Z-disk dysfunction.

Humans↗