3D Structure Prediction Research Articles

Accurate RNA 3D structure prediction using a language model-based deep learning approach.

Accurate prediction of RNA three-dimensional (3D) structures remains an unsolved challenge. Determining RNA 3D structures is crucial for understanding their functions and informing RNA-targeting drug development and synthetic biology design. The structural flexibility of RNA, which leads to the scarcity of experimentally determined data, complicates computational prediction efforts. Here we present RhoFold+, an RNA language model-based deep learning method that accurately predicts 3D structures of single-chain RNAs from sequences. By integrating an RNA language model pretrained on ~23.7 million RNA sequences and leveraging techniques to address data scarcity, RhoFold+ offers a fully automated end-to-end pipeline for RNA 3D structure prediction. Retrospective evaluations on RNA-Puzzles and CASP15 natural RNA targets demonstrate the superiority of RhoFold+ over existing methods, including human expert groups. Its efficacy and generalizability are further validated through cross-family and cross-type assessments, as well as time-censored benchmarks. Additionally, RhoFold+ predicts RNA secondary structures and interhelical angles, providing empirically verifiable features that broaden its applicability to RNA structure and function studies.

Unraveling the physiochemical characteristics and molecular insights of Zein protein through structural modeling and conformational dynamics: a synergistic approach between machine learning and molecular dynamics simulations

This research article presents a comprehensive investigation into the three-dimensional structure, physicochemical characteristics and conformational stability of the Zein protein. Machine learning (ML) based homology modeling approach, was employed to predict the 3D structure of Zein protein. Convolutional neural networks (CNNs) were utilized for refining the model, capturing complex spatial features and improving decoy refinement. The predicted 3D structure of Zein protein showed a high-confidence score, i.e. C-score of 0.96. Physiochemical characteristic was also analyzed to investigate its protonation and deprotonation behavior across a range of pH values. A comprehensive analysis of the titration curve and electrostatic charges was performed to uncover valuable molecular insights into the zein protein’s charge distribution, electrostatic interactions and potential conformational changes. Molecular dynamics (MD) simulations were performed to analyze the zein structural behavior under different pH values (2.0, 4.5, 6.8, 10.0 and 12.5), ionic strengths (0 mM, 25 mM, 50 mM, 75 mM, 100 mM) and temperatures (300K, 350K, 375K). Our results demonstrated the influence of these factors on zein protein’s stability and conformational dynamics. At extreme pH values of 2.0 and 12.5, the Zein protein exhibited increased structural deviations and potential unfolding, while intermediate pH values closer to the protein’s isoelectric point (pI) demonstrated more compact and stable conformations. Analysis of root mean square deviation, radius of gyration, solvent accessible surface area and Ramachandran plot provided clear understandings of the protein’s compactness and surface exposure, confirming the impact of pH, ionic strength and temperature on the protein’s conformation.

IsRNAcirc: 3D structure prediction of circular RNAs based on coarse-grained molecular dynamics simulation.

As an emerging class of RNA molecules, circular RNAs play pivotal roles in various biological processes, thereby determining their three-dimensional (3D) structure is crucial for a deep understanding of their biological significances. Similar to linear RNAs, the development of computational methods for circular RNA 3D structure prediction is challenging, especially considering the inherent flexibility and potentially long length of circular RNAs. Here, we introduce an extension of our previous IsRNA2 model, named IsRNAcirc, to enable circular RNA 3D structure predictions through coarse-grained molecular dynamics simulations. The workflow of IsRNAcirc consists of four main steps, including input preparation, end closure, structure prediction, and model refinement. Our results demonstrate that IsRNAcirc can provide reasonable 3D structure predictions for circular RNAs, which significantly reduce the locally irrational elements contained in the initial input. Moreover, for a validation test set comprising 34 circular RNAs, our IsRNAcirc can generate 3D models with better scores than the template-based 3dRNA method. These findings demonstrate that our IsRNAcirc method is a promising tool to explore the structural details along with intricate interactions of circular RNAs.

Computational-guided discovery of UDP-glycosyltransferases for lauryl glucoside production using engineered E. coli

Defining suitable enzymes for reaction steps in novel synthetic pathways is crucial for developing microbial cell factories for non-natural products. Here, we developed a computational workflow to identify C12 alcohol-active UDP-glycosyltransferases. The workflow involved three steps: (1) assembling initial candidates of putative UDP-glycosyltransferases, (2) refining selection by examining conserved regions, and (3) 3D structure prediction and molecular docking. Genomic sequences from Candida, Pichia, Rhizopus, and Thermotoga, known for lauryl glucoside synthesis via whole-cell biocatalysis, were screened. Out of 240 predicted glycosyltransferases, 8 candidates annotated as glycosyltransferases were selected after filtering out those with signal peptides and identifying conserved UDP-glycosyltransferase regions. These proteins underwent 3D structure prediction and molecular docking with 1-dodecanol. RO3G, a candidate from Rhizopus delemar RA 99–880 with a relatively high ChemPLP fitness score, was selected and expressed in Escherichia coli BL21 (DE3). It was further characterized using a feeding experiment with 1-dodecanol. Results confirmed that the RO3G-expressing strain could convert 1-dodecanol to lauryl glucoside, as quantified by HPLC and identified by targeted LC-MS. Monitoring the growth and fermentation profiles of the engineered strain revealed that RO3G expression did not affect cell growth. Interestingly, acetate, a major fermentation product, was reduced in the RO3G-expressing strain compared to the GFP-expressing strain, suggesting a redirection of flux from acetate to other pathways. Overall, this work presents a successful workflow for discovering UDP-glycosyltransferase enzymes with confirmed activity toward 1-dodecanol for lauryl glucoside production.Graphical abstract

Design and development of dual targeting CAR protein for the development of CAR T-cell therapy against KRAS mutated pancreatic ductal adenocarcinoma using computational approaches

Mutant KRAS promotes the proliferation, metastasis, and aggressiveness of various cancers including pancreatic ductal adenocarcinoma (PDAC), non-small cell lung cancer (NSCLC), and colorectal adenocarcinoma (CRC) respectively. Mutant KRAS therapeutics are limited, while Sotorasib and Adagrasib were the only FDA-approved drugs for the treatment of KRASG12C mutated NSCLC. Chimeric antigen receptor (CAR) T-cell therapy has been emerged as an effective strategy against hematological malignancies and being extended towards solid cancers including PDAC. mesothelin (MSLN) and Carcinoembryonic Antigen (CEA) were reported to be highly overexpressed in KRAS-mutated PDAC. Meanwhile, in clinical trials, several CAR T-cell therapy studies are mainly focused towards these two cancer antigens in PDAC, however, the dual targeting of these two neoantigens is not reported. In the present study, we have designed and developed a novel dual-targeting CAR protein by employing various bioinformatics approaches such as functional analysis (antigenicity, allergenicity, antigen binding sites & signalling cascades), qualitative analysis (physicochemical, prediction, refinement & validation of 2D and 3D structures), molecular docking, and in silico cloning. Our results revealed that the designed CAR protein specifically binds with both MSLN & CEA with significant binding affinities, and was predicted to be stable & non-allergenic. Additionally, the protein–protein interaction network reveals the T-cell mediated antitumor responses of each domain in the designed CAR. Conclusively, we have designed and developed a dual targeting (MSLN & CEA) CAR protein towards KRAS-mutated PDAC using computational approaches. Alongside, we further recommend to engineer this designed CAR in T-cells and evaluating their therapeutic efficiency in in vitro and in vivo studies in the near future.

Dinucleotide composition representation -based deep learning to predict scoliosis-associated Fibrillin-1 genotypes.

Scoliosis is a pathological spine structure deformation, predominantly classified as "idiopathic" due to its unknown etiology. However, it has been suggested that scoliosis may be linked to polygenic backgrounds. It is crucial to identify potential Adolescent Idiopathic Scoliosis (AIS)-related genetic backgrounds before scoliosis onset. The present study was designed to intelligently parse, decompose and predict AIS-related variants in ClinVar database. Possible AIS-related variant records downloaded from ClinVar were parsed for various labels, decomposed for Dinucleotide Compositional Representation (DCR) and other traits, screened for high-risk genes with statistical analysis, and then learned intelligently with deep learning to predict high-risk AIS genotypes. Results demonstrated that the present framework is composed of all technical sections of data parsing, scoliosis genotyping, genome encoding, machine learning (ML)/deep learning (DL) and scoliosis genotype predicting. 58,000 scoliosis-related records were automatically parsed and statistically analyzed for high-risk genes and genotypes, such as FBN1, LAMA2 and SPG11. All variant genes were decomposed for DCR and other traits. Unsupervised ML indicated marked inter-group separation and intra-group clustering of the DCR of FBN1, LAMA2 or SPG11 for the five types of variants (Pathogenic, Pathogeniclikely, Benign, Benignlikely and Uncertain). A FBN1 DCR-based Convolutional Neural Network (CNN) was trained for Pathogenic and Benign/ Benignlikely variants performed accurately on validation data and predicted 179 high-risk scoliosis variants. The trained predictor was interpretable for the similar distribution of variant types and variant locations within 2D structure units in the predicted 3D structure of FBN1. In summary, scoliosis risk is predictable by deep learning based on genomic decomposed features of DCR. DCR-based classifier has predicted more scoliosis risk FBN1 variants in ClinVar database. DCR-based models would be promising for genotype-to-phenotype prediction for more disease types.

The Structure Analysis and mRNA Expression of CaV2 Gene Responding to Hypoxia Stress in Anadara granosa

The blood clam (Anadara granosa) is an economic bivalve that is relatively tolerant to hypoxia, but its molecular mechanism of hypoxia tolerance is unclear. We found that a significant decrease in extracellular Ca2+ concentration and a marked increase in intracellular Ca2+ concentration was observed in the blood clam through the fluorescence probe method, under hypoxic conditions at 0.5 mg/L. Concomitantly, there was a downward trend in the expression level of CaV2 mRNA, whereas NFAT (nuclear factor of activated T cells) expression increased by qRT-PCR. These findings suggest that the elevated intracellular Ca2+ concentration may activate negative transcription factors of NFAT, which subsequently suppresses the transcription of CaV2, leading to its decreased expression. Then, the NFAT RNA interference experiments supported this hypothesis. Sequence analysis and 3D structure prediction revealed conserved and mutated residue sites in blood clam compared to other bivalves. Hypoxia-induced changes in intracellular and extracellular Ca2+ concentrations, activating transcription factor NFAT and suppressing CaV2 expression. This study highlights the key roles of CaV2 and NFAT in hypoxia adaptation, paving the way for further exploration of hypoxia tolerance mechanisms in mollusca.

Experimental Spectroscopic and Computational Studies on A New Synthesized Sulfisoxazole Derivative; Molecular Docking, Drug-likeness, ADME, Toxicity Predictions and Carbonic Anhydrase II Activity Investigations

In this paper, 5-Dimethylamino-2-hydroxy-3-methoxy benzaldehyde sulfisoxazole (5DHMS) and Cu(5DHMS)2 compounds have been synthesized. The molecular structure characterizations were performed using experimental and computational methods. In the experimental part of the study, CHNS, FT-IR, NMR, TGA, and LC-MS analysis techniques were used. In the theoretical part, Density Functional Theory (DFT) was selected for the calculation level. Firstly, optimized structures were obtained from the predicted 3D structures. Vibrational modes were calculated using the optimized structures of the compounds. The vibrational modes of each compound were analyzed in detail using potential energy distribution (PED). Molecular electrostatic potential and HOMO-LUMO maps were drawn. Global chemical reactivity descriptors of compounds were determined. Moreover, the solvent media effects on chemical reactivity descriptors were revealed in detail. Protein-ligand interactions with the target receptor carbonic anhydrase II enzyme were performed. In addition, some important biological activity parameters such as drug-likeness, toxicity, and ADME predictions were examined in silico.

Computational Analysis and Inhibition Study of Carbonic Anhydrase from Azadirachta indica Leaves

Introduction: Carbonic anhydrase (CA) has been an enzyme of great interest for its application in carbon dioxide sequestration. A total of 66 protein sequences of plant CAs were computationally analyzed and submitted to bioinformatics for motif and domain identification, multiple sequence alignment, and phylogenetic analysis. Method: The current work aims to extend knowledge among researchers in better understanding the structure of AZDI CA by analyzing its physicochemical properties, secondary structure prediction, 3D modeling of protein sequence, and its validation using a variety of conventional computational methods. Results: The accuracy of the predicted 3D structure checked by the Ramachandran plot generated by PROCHECK showed that for the CA protein sequence 93.1%, was observed in the most favored regions, VERIFY 3D 69.44% and ERRAT 98.09%. AZDI CA was docked with various anions and sulphonamide derivatives to study their inhibitory effect and calculate binding energy by using Autodock 4.2. Ethoxzolamide is found to be a better inhibitor of AZDI CA with binding energy -8.71 kCal/mol and KI value 0.0004 mM. Further MD simulation of the docked complex of ethoxzolamide with AZDI CA was done using SCHRODINGER DESMOND software. Conclusion: The outcomes of this research will have a significant influence on biochemistry, biotechnology, and potential applications of AZDI CA and similar plant CAs in determining potent inhibitors for drug targets in treating various diseases.

In silico Discovery of Leptukalins, The New Potassium Channel Blockers from the Iranian Scorpion, Hemiscorpius Lepturus.

Blocking Kv 1.2 and Kv 1.3 potassium channels using scorpion venom- derived toxins holds potential therapeutic value. These channels are implicated in autoimmune diseases such as neurodegenerative diseases, multiple sclerosis, rheumatoid arthritis, and type 1 diabetes. The present work aims at the discovery and in silico activity analysis of potassium channel blockers (KTxs) from the cDNA library derived from the venom gland of Iranian scorpion Hemiscorpius lepturus (H. lepturus). The sequence regarding potassium channel blockers were extracted based on Gene Ontology for H. lepturus venom gland. Homology analyses, superfamily, family, and evolutionary signatures of H. lepturus KTxs (H.L KTxs) were determined by using BLASTP, COBALT, PROSITE, and InterPro servers. The predicted 3D structures of H.L KTxs were superimposed against their homologs to predict structure activity relationship. Molecular docking analysis was also performed to predict the binding affinity of H.L KTxs to Kv 1.2 and Kv 1.3 channels. Finally, the toxicity was predicted. Seven H.L KTxs, designated as Leptukalin, were extracted from the cDNA library of H. lepturus venom gland. Homology analyses proved that they can act as potassium channel blockers and they belong to the superfamily and family of Scorpion Toxin-like and Short-chain scorpion toxins, respectively. Structural alignment results confirmed the activity of H.L KTxs. Binding affinity of all H.L KTxs to Kv 1.2 and Kv 1.3 channels ranged from -4.4 to -5.5 and -4 to -5.7 Kcal/mol, respectively. In silico toxicity assay showed that Leptukalin 3, Leptukalin 5, and Leptukalin 7 were non-toxic. Three non-toxic KTxs, Leptukalin 3, 5, and 7, were successfully discovered from the cDNA library of H. lepturus venom gland. Gathering all data together, the discovered peptides are promising potassium channel blockers. Accordingly, Leptukalin 3, 5, and 7 could be suggested for complementary in vitro studies and mouse model of autoimmune 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.

Analysis of Phylogenetics and Structural Relationships of the Cytochrome b Gene Protein in Mahseer Fish (Tor sp.)

Abstract The Cytochrome b gene from mitochondrial DNA encodes a protein that can be used as a genetic marker to identify species. This study aims to determine the bioinformatic characteristics of the gene encoding Cyt b isolated from various Mahseer species, Tor tambroides, Tor douronensis, Tor khudree, and Tor putitora. Mahseer Cyt b sequences can be used for detailed computational investigation of the properties of 3D protein models and phylogenetics. The data analysis technique uses nucleotides with a sequence length of 330 bp obtained from NCBI. The MEGA X program was used for phylogenetic analysis. 3D protein structure predictions were obtained through the Swiss model and protein homology program using PYMOL software. The results of phylogenetics show close kinship relationships with the closest genetic distance of 0.000, while the farthest distance is 0.068. The nucleotide composition contains Thymine (30.23%), Cytosine (25.81%), Arginine (28.05%), and Guanine (15.92%). 3D modeling of the Cyt b protein Mahseer referring to the species Neolissochilus hexagonolepis (McClelland, 1839). The 3D protein structure prediction was assessed via a Ramachandran plot of 99.07%. This study can be used as a reference to support sustainable conservation efforts by observing the populations and habitat conditions.

Development of Drug Discovery Platforms Using Artificial Intelligence and Cheminformatics.

Recently, remarkable progress has been achieved in artificial intelligence (AI), including machine learning. Various AI models have been proposed for drug discovery, including the design of small molecules, activity prediction, and three-dimensional (3D) structure prediction of proteins. AI consists of diverse elements, including information retrieval and machine learning, and can be used in a wide range of drug discovery scenarios. In this review, we focused on AI for small-molecule drug discovery with respect to molecular design, activity prediction, and prediction of the binding poses of compounds to target molecules. We also discussed the applications of AI in academic drug discovery.

Intracellular bactericidal activity and action mechanism of MDP1 antimicrobial peptide against VRSA and MRSA in human endothelial cells.

Staphylococcus aureus is a prominent cause of postoperative infections, often persisting within host cells, leading to chronic infections. Conventional antibiotics struggle to eliminate intracellular S. aureus due to poor cell penetration. Antimicrobial peptides are a new hope for tackling intracellular bacteria. Accordingly, this study examines the antimicrobial peptide MDP1, derived from melittin, for its efficacy against intracellular S. aureus. In this study, the physiochemical properties (Prediction of three-dimensional structure, circular dichroism and helical wheel projection analysis) were investigated. Extracellular antibacterial activity and cytotoxicity of MDP1 were also assessed. The mechanism of interaction of MDP1 with S. aureus was evaluated by molecular dynamic simulation, atomic force and confocal microscopy. Bacterial internalization into an endothelial cell model was confirmed through culture and transmission electron microscopy. The effect of the peptide on intracellular bacteria was investigated by culture and epi-fluorescence microscopy. 3D structural prediction proved the conformation of MDP1 as an α-helix peptide. Helical-wheel projection analysis indicated the proper orientation of hydrophobic amino acid residues for membrane interaction. CD spectroscopy of MDP1 showed that MDP1 in SDS 10 and 30 mM adopted 87 and 91% helical conformation. Atomic force and confocal microscopy assessments as well as molecular dynamics studies revealed the peptide-bacterial membrane interaction. MDP1, at the concentration of 0.32 μg mL-1, demonstrated a fold reduction of 21.7 ± 1.8, 1.7 ± 0.2, and 7.3 ± 0.8 in intracellular bacterial load for ATCC, VRSA, and MRSA, respectively. Molecular dynamics results demonstrate a preferential interaction of MDP1 with POPG/POPE membranes, primarily driven by electrostatic forces and hydrogen bonding. In POPC systems, two out of four MDP1 interacted effectively, while all four MDP1 engaged with POPG/POPE membranes. Gathering all data together, MDP1 is efficacious in the reduction of intracellular VRSA and MRSA proved by culture and epi-fluorescent microscopy although further studies should be performed to increase the intracellular activity of MDP1.

Artificial intelligence in fusion protein three-dimensional structure prediction: Review and perspective.

Recent advancements in artificial intelligence (AI) have accelerated the prediction of unknown protein structures. However, accurately predicting the three-dimensional (3D) structures of fusion proteins remains a difficult task because the current AI-based protein structure predictions are focused on the WT proteins rather than on the newly fused proteins in nature. Following the central dogma of biology, fusion proteins are translated from fusion transcripts, which are made by transcribing the fusion genes between two different loci through the chromosomal rearrangements in cancer. Accurately predicting the 3D structures of fusion proteins is important for understanding the functional roles and mechanisms of action of new chimeric proteins. However, predicting their 3D structure using a template-based model is challenging because known template structures are often unavailable in databases. Deep learning (DL) models that utilize multi-level protein information have revolutionized the prediction of protein 3D structures. In this review paper, we highlighted the latest advancements and ongoing challenges in predicting the 3D structure of fusion proteins using DL models. We aim to explore both the advantages and challenges of employing AlphaFold2, RoseTTAFold, tr-Rosetta and D-I-TASSER for modelling the 3D structures. HIGHLIGHTS: This review provides the overall pipeline and landscape of the prediction of the 3D structure of fusion protein. This review provides the factors that should be considered in predicting the 3D structures of fusion proteins using AI approaches in each step. This review highlights the latest advancements and ongoing challenges in predicting the 3D structure of fusion proteins using deep learning models. This review explores the advantages and challenges of employing AlphaFold2, RoseTTAFold, tr-Rosetta, and D-I-TASSER to model 3D structures.

In silico functional analysis of DNA repair proteins (OGG1, MYH) in Arabidopsis thaliana

Deoxyribonucleic acid (DNA) is a hereditary material found in every living organism. However, every organism is exposed to endogenous and exogenous an agent which causes damage to DNA molecules. Even though genetic variation is important for evolution, every organism requires genetic stability for survival. Not only it is needed for DNA to replicate accurately but also several DNA repair mechanisms are used to fix the damage which occurred in DNA. A detailed structure analysis along with docking site and metabolic pathway prediction were used to confirm that these proteins are involved in DNA repair mechanisms. This study investigated OGG1 and MYH proteins from Arabidopsis thaliana, and showed their interaction with predicted partners, focusing on their detailed interactome and 3D structure predictions. This study concluded that OGG1 and MYH are not only found in DNA repair mechanisms but also show strong functions in DNA replication and the development of plant and cell cycle.

Evaluating the 3D structure prediction tools to identify optimal MEBPVC structure models

Multi-Epitope based Peptide Vaccines Candidates (MEBPVCs) are peptide sequences with the immunogenic epitopes interspersed with linkers, adjuvants and other components. The optimization of one of the components of MEBPVCs, epitope order, resulted in a unique epitope ordered MEBPVC showing improved vaccine potency. In addition to the above combinatorial optimization task, predicting the 3D structure of these MEBPVCs presents a significant challenge owing to the diverse modeling approaches employed by the modeling tools. Here we embarked on evaluating the performance of the threading based ab initio modeling tool: I-TASSER, Large Language Model (LLM) based tool: ESMFold and the MSA based artificial intelligence based tool: AlphaFold2. Our investigation employed RMSD, PEA, SASA, Solubility, Globularity, secondary structure analysis, Docking, MDS and MVP. I-TASSER emerged as the superior performer. The analysis revealed a strong correlation between predicted tertiary structure and secondary structure composition, especially the β-turns and random coils. It appears that accurate secondary structure identification to be a pivotal step, dictating the propagation of accuracy (or inaccuracy). This finding highlights the critical importance of secondary structure prediction/assignment algorithms for enhanced protein structure modeling accuracy especially to integrate predictions where evolutionary information is minimal and or lacking. Small sample size is attributed to the low significance of the parameters though Globularity and SASA have shown good p-values. Overall, the results support the need and utilization of the LLMs in biological research, underscoring their potential to revolutionize protein structure prediction and advance diverse biomolecular applications, accelerating Synthetic Biology research.

The Impact of Vp-Porin, an Outer Membrane Protein, on the Biological Characteristics and Virulence of Vibrio Parahaemolyticus.

Porins are crucial proteins located in the outer membrane that directly influence antimicrobial resistance mechanisms and virulence in bacteria. In this study, a porin gene (Vp-porin) was cloned in V. parahaemolyticus, and the function of Vp-Porin in biological characteristics and virulence was investigated. The results of sequence analysis showed that Vp-Porin is highly conserved in Vibrio spp., and the predicted 3D structure showed it could form a 20-strand transmembrane β-barrel domian. Membrane permeabilization provides evidence that the membrane integrity of ∆Vp-porin was damaged and the sensitivity to tetracycline, polymyxin B, rifampicin and cephalothin of ∆Vp-porin obviously increased. In addition, loss of Vp-porin damaged motility due to downregulated flagellar synthesis. In addition, ∆Vp-porin exhibited attenuated cytotoxicity to Tetrahymena. The relative survival rate of Tetrahymena infection with ∆Vp-porin was 86%, which is much higher than that with WT (49%). Taken together, the results of this study indicate that Vp-Porin in V. parahaemolyticus plays various roles in biological characteristics in membrane integrity, antimicrobial resistance and motility and contributes to virulence.

Comparative analysis of RNA 3D structure prediction methods: towards enhanced modeling of RNA-ligand interactions.

Accurate RNA structure models are crucial for designing small molecule ligands that modulate their functions. This study assesses six standalone RNA 3D structure prediction methods-DeepFoldRNA, RhoFold, BRiQ, FARFAR2, SimRNAand Vfold2, excluding web-based tools due to intellectual property concerns. We focus on reproducing the RNA structure existing in RNA-small molecule complexes, particularly on the ability to model ligand binding sites. Using a comprehensive set of RNA structures from the PDB, which includes diverse structural elements, we found that machine learning (ML)-based methods effectively predict global RNA folds but are less accurate with local interactions. Conversely, non-ML-based methods demonstrate higher precision in modeling intramolecular interactions, particularly with secondary structure restraints. Importantly, ligand-binding site accuracy can remain sufficiently high for practical use, even if the overall model quality is not optimal. With the recent release of AlphaFold 3, we included this advanced method in our tests. Benchmark subsets containing new structures, not used in the training of the tested ML methods, show that AlphaFold 3's performance was comparable to other ML-based methods, albeit with some challenges in accurately modeling ligand binding sites. This study underscores the importance of enhancing binding site prediction accuracy and the challenges in modeling RNA-ligand interactions accurately.

APACE: AlphaFold2 and advanced computing as a service for accelerated discovery in biophysics

The prediction of protein 3D structure from amino acid sequence is a computational grand challenge in biophysics and plays a key role in robust protein structure prediction algorithms, from drug discovery to genome interpretation. The advent of AI models, such as AlphaFold, is revolutionizing applications that depend on robust protein structure prediction algorithms. To maximize the impact, and ease the usability, of these AI tools we introduce APACE, AlphaFold2 and advanced computing as a service, a computational framework that effectively handles this AI model and its TB-size database to conduct accelerated protein structure prediction analyses in modern supercomputing environments. We deployed APACE in the Delta and Polaris supercomputers and quantified its performance for accurate protein structure predictions using four exemplar proteins: 6AWO, 6OAN, 7MEZ, and 6D6U. Using up to 300 ensembles, distributed across 200 NVIDIA A100 GPUs, we found that APACE is up to two orders of magnitude faster than off-the-self AlphaFold2 implementations, reducing time-to-solution from weeks to minutes. This computational approach may be readily linked with robotics laboratories to automate and accelerate scientific discovery.