Structure Prediction Research Articles

AlphaFold-latest: revolutionizing protein structure prediction for comprehensive biomolecular insights and therapeutic advancements

AlphaFold-latest: revolutionizing protein structure prediction for comprehensive biomolecular insights and therapeutic advancements

The Rapid Evolution of De Novo Proteins in Structure and Complex.

Recent studies in the rice genome-wide have established that de novo genes, evolving from non-coding sequences, enhance protein diversity through a stepwise process. However, the pattern and rate of their evolution in protein structure over time remain unclear. Here, we addressed these issues within a surprisingly short evolutionary timescale (<1 million years for 97% of Oryza de novo genes) with comparative approaches to gene duplicates. We found that de novo genes evolve faster than gene duplicates in the intrinsic disordered regions (IDRs, such as random coils), secondary structure elements (such as α-helix and β-strand), hydrophobicity, and molecular recognition features (MoRFs). In de novo proteins, specifically, we observed an 8-14% decay in random coils and IDR lengths and a 2.3-6.5% increase in structured elements, hydrophobicity, and MoRFs, per million years on average. These patterns of structural evolution align with changes in amino acid composition over time as well. We also revealed higher positive charges but smaller molecular weights for de novo proteins than duplicates. Tertiary structure predictions showed that most de novo proteins, though not typically well-folded on their own, readily form low-energy and compact complexes with other proteins facilitated by extensive residue contacts and conformational flexibility, suggesting a faster-binding scenario in de novo proteins to promote interaction. These analyses illuminate a rapid evolution of protein structure in de novo genes in rice genomes, originating from non-coding sequences, highlighting their quick transformation into active, protein complex-forming components within a remarkably short evolutionary timeframe.

Exploration of three Dyadobacter fermentans enzymes uncovers molecular activity determinants in CE15

Glucuronoyl esterases (GEs) are serine-type hydrolase enzymes belonging to carbohydrate esterase family 15 (CE15), and they play a central role in the reduction of recalcitrance in plant cell walls by cleaving ester linkages between glucuronoxylan and lignin in lignocellulose. Recent studies have suggested that bacterial CE15 enzymes are more heterogeneous in terms of sequence, structure, and substrate preferences than their fungal counterparts. However, the sequence space of bacterial GEs has still not been fully explored, and further studies on diverse enzymes could provide novel insights into new catalysts of biotechnological interest. To expand our knowledge on this family of enzymes, we investigated three unique CE15 members encoded by Dyadobacter fermentans NS114T, a Gram-negative bacterium found endophytically in maize/corn (Zea mays). The enzymes are dissimilar, sharing ≤ 39% sequence identity to each other‚ and were considerably different in their activities towards synthetic substrates. Combined analysis of their primary sequences and structural predictions aided in establishing hypotheses regarding specificity determinants within CE15, and these were tested using enzyme variants attempting to shift the activity profiles. Together, the results expand our existing knowledge of CE15, shed light into the molecular determinants defining specificity, and support the recent thesis that diverse GEs encoded by a single microorganism may have evolved to fulfil different physiological functions.Key points• D. fermentans encodes three CE15 enzymes with diverse sequences and specificities• The Region 2 inserts in bacterial GEs may directly influence enzyme activity• Rational amino acid substitutions improved the poor activity of the DfCE15A enzyme

Mapping the Intersubunit Interdomain FMN-Heme Interactions in Neuronal Nitric Oxide Synthase by Targeted Quantitative Cross-Linking Mass Spectrometry.

Nitric oxide synthase (NOS) in mammals is a family of multidomain proteins in which interdomain electron transfer (IET) is controlled by domain-domain interactions. Calmodulin (CaM) binds to the canonical CaM-binding site in the linker region between the FMN and heme domains of NOS and allows tethered FMN domain motions, enabling an intersubunit FMN-heme IET in the output state for NO production. Our previous cross-linking mass spectrometric (XL MS) results demonstrated site-specific protein dynamics in the CaM-responsive regions of rat neuronal NOS (nNOS) reductase construct, a monomeric protein [Jiang et al., Biochemistry, 2023, 62, 2232-2237]. In this work, we have extended our combined approach of XL MS structural mapping and AlphaFold structural prediction to examine the homodimeric nNOS oxygenase/FMN (oxyFMN) construct, an established model of the NOS output state. We employed parallel reaction monitoring (PRM) based quantitative XL MS (qXL MS) to assess the CaM-induced changes in interdomain dynamics and interactions. Intersubunit cross-links were identified by mapping the cross-links onto top AlphaFold structural models, which was complemented by comparing their relative abundances in the cross-linked dimeric and monomeric bands. Furthermore, contrasting the CaM-free and CaM-bound nNOS samples shows that CaM enables the formation of the intersubunit FMN-heme docking complex and that CaM binding induces extensive, allosteric conformational changes across the NOS regions. Moreover, the observed cross-links sites specifically respond to changes in ionic strength. This indicates that interdomain salt bridges are responsible for stabilizing and orienting the output state for efficient FMN-heme IET. Taken together, our targeted qXL MS results have revealed that CaM and ionic strength modulate specific dynamic changes in the CaM/FMN/heme complexes, particularly in the context of intersubunit interdomain FMN-heme interactions.

Phylum-level studies of bacterial cutinases for unravelling enzymatic specificity toward PET degradation: an in silico approach.

The overwhelming use of PET plastic in various day-to-day activities led to the voluminous increase in PET waste and growing environmental hazards. A plethora of methods have been used that are associated with secondary pollutants. Therefore, microbial degradation of PET provides a sustainable approach due to its versatile metabolic diversity and capacity. The present work highlights the cutinase enzyme's role in PET degradation. This study focuses on the bacterial cutinases homologs screened from 43 reported phylum of bacteria. The reported bacterial cutinases for plastic degradation have been chosen as reference sequences, and 917 sequences have shown homology across the bacterial phyla. The dienelactone hydrolase (DLH) domain was identified for attaining specificity towards PET binding in 196 of 917 sequences. Various computational tools have been used for the physicochemical characterization of 196 sequences. The analysis revealed that most selected sequences are hydrophilic, extracellular, and thermally stable. Based on this analysis, 17 sequences have been further pursued for three-dimensional structure prediction and validation. The molecular docking studies of 17 selected sequences revealed efficient PET binding with the three sequences derived from the phylum Bacteroidota, the lowest binding energy of -5.9kcal/mol, Armatimonadota, and Nitrososphaerota with -5.8kcal/mol. The two enzyme sequences retrieved from the phylum Bacteroidota and Armatimonadota are metagenomically derived. Therefore, the present studies concluded that there is a high probability of finding cutinase homologs in various environmental resources that can be further explored for PET degradation.

Prediction of specific structural damage to the knee joint using qualitative isokinetic analysis

BackgroundAn isokinetic moment curve (IMC) pattern-damaged structure prediction model may be of considerable value in assisting the diagnosis of knee injuries in clinical scenarios. This study aimed to explore the association between irregular IMC patterns and specific structural damages in the knee, including anterior cruciate ligament (ACL) rupture, meniscus (MS) injury, and patellofemoral joint (PFJ) lesions, and to develop an IMC pattern-damaged structure prediction model.MethodsA total of 94 subjects were enrolled in this study and underwent isokinetic testing of the knee joint (5 consecutive flexion-extension movements within the range of motion of 90°-10°, 60°/s). Qualitative analysis of the IMCs for all subjects was completed by two blinded examiners. A multinomial logistic regression analysis was used to investigate whether a specific abnormal curve pattern was associated with specific knee structural injuries and to test the predictive effectiveness of IMC patterns for specific structural damage in the knee.ResultsThe results of the multinomial logistic regression revealed a significant association between the irregular IMC patterns of the knee extensors and specific structural damages (“Valley” - ACL, PFJ, and ACL + MS, “Drop” - ACL, and ACL + MS, “Shaking” - ACL, MS, PFJ, and ACL + MS). The accuracy and Macro-averaged F1 score of the predicting model were 56.1% and 0.426, respectively.ConclusionThe associations between irregular IMC patterns and specific knee structural injuries were identified. However, the accuracy and Macro-averaged F1 score of the established predictive model indicated its relatively low predictive efficacy. For the development of a more accurate predictive model, it may be essential to incorporate angle-specific and/or speed-specific analyses of qualitative and quantitative data in isokinetic testing. Furthermore, the utilization of artificial intelligence image recognition technology may prove beneficial for analyzing large datasets in the future.

OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization.

AlphaFold2 revolutionized structural biology with the ability to predict protein structures with exceptionally high accuracy. Its implementation, however, lacks the code and data required to train new models. These are necessary to (1) tackle new tasks, like protein-ligand complex structure prediction, (2) investigate the process by which the model learns and (3) assess the model's capacity to generalize to unseen regions of fold space. Here we report OpenFold, a fast, memory efficient and trainable implementation of AlphaFold2. We train OpenFold from scratch, matching the accuracy of AlphaFold2. Having established parity, we find that OpenFold is remarkably robust at generalizing even when the size and diversity of its training set is deliberately limited, including near-complete elisions of classes of secondary structure elements. By analyzing intermediate structures produced during training, we also gain insights into the hierarchical manner in which OpenFold learns to fold. In sum, our studies demonstrate the power and utility of OpenFold, which we believe will prove to be a crucial resource for the protein modeling community.

The big five factors as differential predictors of self-regulation, achievement emotions, coping and health behavior in undergraduate students

BackgroundThe aim of this research was to analyze whether the personality factors included in the Big Five model differentially predict the self-regulation and affective states of university students and health.MethodsA total of 637 students completed validated self-report questionnaires. Using an ex post facto design, we conducted linear regression and structural prediction analyses.ResultsThe findings showed that model factors were differential predictors of both self-regulation and affective states. Self-regulation and affective states, in turn, jointly predict emotional performance while learning and even student health. These results allow us to understand, through a holistic predictive model, the differential predictive relationships of all the factors: conscientiousness and extraversion were predictors regulating positive emotionality and health; the openness to experience factor was non-regulating; nonregulating; and agreeableness and neuroticism were dysregulating, hence precursors of negative emotionality and poorer student health.ConclusionsThese results are important because they allow us to infer implications for guidance and psychological health at university.

Multi-purpose RNA language modelling with motif-aware pretraining and type-guided fine-tuning

AbstractPretrained language models have shown promise in analysing nucleotide sequences, yet a versatile model excelling across diverse tasks with a single pretrained weight set remains elusive. Here we introduce RNAErnie, an RNA-focused pretrained model built upon the transformer architecture, employing two simple yet effective strategies. First, RNAErnie enhances pretraining by incorporating RNA motifs as biological priors and introducing motif-level random masking in addition to masked language modelling at base/subsequence levels. It also tokenizes RNA types (for example, miRNA, lnRNA) as stop words, appending them to sequences during pretraining. Second, subject to out-of-distribution tasks with RNA sequences not seen during the pretraining phase, RNAErnie proposes a type-guided fine-tuning strategy that first predicts possible RNA types using an RNA sequence and then appends the predicted type to the tail of sequence to refine feature embedding in a post hoc way. Our extensive evaluation across seven datasets and five tasks demonstrates the superiority of RNAErnie in both supervised and unsupervised learning. It surpasses baselines with up to 1.8% higher accuracy in classification, 2.2% greater accuracy in interaction prediction and 3.3% improved F1 score in structure prediction, showcasing its robustness and adaptability with a unified pretrained foundation.

Predicting Antimicrobial Peptides Using ESMFold-Predicted Structures and ESM-2-Based Amino Acid Features with Graph Deep Learning.

Currently, antimicrobial resistance constitutes a serious threat to human health. Drugs based on antimicrobial peptides (AMPs) constitute one of the alternatives to address it. Shallow and deep learning (DL)-based models have mainly been built from amino acid sequences to predict AMPs. Recent advances in tertiary (3D) structure prediction have opened new opportunities in this field. In this sense, models based on graphs derived from predicted peptide structures have recently been proposed. However, these models are not in correspondence with state-of-the-art approaches to codify evolutionary information, and, in addition, they are memory- and time-consuming because depend on multiple sequence alignment. Herein, we presented a framework to create alignment-free models based on graph representations generated from ESMFold-predicted peptide structures, whose nodes are characterized with amino acid-level evolutionary information derived from the Evolutionary Scale Modeling (ESM-2) models. A graph attention network (GAT) was implemented to assess the usefulness of the framework in the AMP classification. To this end, a set comprised of 67,058 peptides was used. It was demonstrated that the proposed methodology allowed to build GAT models with generalization abilities consistently better than 20 state-of-the-art non-DL-based and DL-based models. The best GAT models were developed using evolutionary information derived from the 36- and 33-layer ESM-2 models. Similarity studies showed that the best-built GAT models codified different chemical spaces, and thus they were fused to significantly improve the classification. In general, the results suggest that esm-AxP-GDL is a promissory tool to develop good, structure-dependent, and alignment-free models that can be successfully applied in the screening of large data sets. This framework should not only be useful to classify AMPs but also for modeling other peptide and protein activities.

Structural annotation of unknown molecules in a miniaturized mass spectrometer based on a transformer enabled fragment tree method

Structural annotation of small molecules in tandem mass spectrometry has always been a central challenge in mass spectrometry analysis, especially using a miniaturized mass spectrometer for on-site testing. Here, we propose the Transformer enabled Fragment Tree (TeFT) method, which combines various types of fragmentation tree models and a deep learning Transformer module. It is aimed to generate the specific structure of molecules de novo solely from mass spectrometry spectra. The evaluation results on different open-source databases indicated that the proposed model achieved remarkable results in that the majority of molecular structures of compounds in the test can be successfully recognized. Also, the TeFT has been validated on a miniaturized mass spectrometer with low-resolution spectra for 16 flavonoid alcohols, achieving complete structure prediction for 8 substances. Finally, TeFT confirmed the structure of the compound contained in a Chinese medicine substance called the Anweiyang capsule. These results indicate that the TeFT method is suitable for annotating fragmentation peaks with clear fragmentation rules, particularly when applied to on-site mass spectrometry with lower mass resolution.

FBXO11 variants are associated with intellectual disability and variable clinical manifestation in Chinese affected individuals.

F-box protein 11 (FBXO11) is a member of F-Box protein family, which has recently been proved to be associated with intellectual developmental disorder with dysmorphic facies and behavioral abnormalities (IDDFBA, OMIM: 618089). In this study, 12 intellectual disability individuals from 5 Chinese ID families were collected, and whole exome sequencing (WES), sanger sequencing, and RNA sequencing (RNA-seq) were conducted. Almost all the affected individuals presented with mild to severe intellectual disability (12/12), global developmental delay (10/12), speech and language development delay (8/12) associated with a range of alternate features including increased body weight (7/12), short stature (6/12), seizures (3/12), reduced visual acuity (4/12), hypotonia (1/12), and auditory hallucinations and hallucinations (1/12). Distinguishingly, malformation was not observed in all the affected individuals. WES analysis showed 5 novel FBXO11 variants, which include an inframe deletion variant, a missense variant, two frameshift variants, and a partial deletion of FBXO11 (exon 22-23). RNA-seq indicated that exon 22-23 deletion of FBXO11 results in a new mRNA structure. Conservation and protein structure prediction demonstrated deleterious effect of these variants. The DEGs analysis revealed 148 differentially expressed genes shared among 6 affected individuals, which were mainly associated with genes of muscle and immune system. Our research is the first report of FBXO11-associated IDDFBA in Chinese individuals, which expands the genetic and clinical spectrum of this newly identified NDD/ID syndrome.

SimRNAweb v2.0: a web server for RNA folding simulations and 3D structure modeling, with optional restraints and enhanced analysis of folding trajectories.

Research on ribonucleic acid (RNA) structures and functions benefits from easy-to-use tools for computational prediction and analyses of RNA three-dimensional (3D) structure. The SimRNAweb server version 2.0 offers an enhanced, user-friendly platform for RNA 3D structure prediction and analysis of RNA folding trajectories based on the SimRNA method. SimRNA employs a coarse-grained model, Monte Carlo sampling and statistical potentials to explore RNA conformational space, optionally guided by spatial restraints. Recognized for its accuracy in RNA 3D structure prediction in RNA-Puzzles and CASP competitions, SimRNA is particularly useful for incorporating restraints based on experimental data. The new server version introduces performance optimizations and extends user control over simulations and the processing of results. It allows the application of various hard and soft restraints, accommodating alternative structures involving canonical and noncanonical base pairs and unpaired residues, while also integrating data from chemical probing methods. Enhanced features include an improved analysis of folding trajectories, offering advanced clustering options and multiple analyses of the generated trajectories. These updates provide comprehensive tools for detailed RNA structure analysis. SimRNAweb v2.0 significantly broadens the scope of RNA modeling, emphasizing flexibility and user-defined parameter control. The web server is available at https://genesilico.pl/SimRNAweb.

LoCoHD: a metric for comparing local environments of proteins

Protein folds and the local environments they create can be compared using a variety of differently designed measures, such as the root mean squared deviation, the global distance test, the template modeling score or the local distance difference test. Although these measures have proven to be useful for a variety of tasks, each fails to fully incorporate the valuable chemical information inherent to atoms and residues, and considers these only partially and indirectly. Here, we develop the highly flexible local composition Hellinger distance (LoCoHD) metric, which is based on the chemical composition of local residue environments. Using LoCoHD, we analyze the chemical heterogeneity of amino acid environments and identify valines having the most conserved-, and arginines having the most variable chemical environments. We use LoCoHD to investigate structural ensembles, to evaluate critical assessment of structure prediction (CASP) competitors, to compare the results with the local distance difference test (lDDT) scoring system, and to evaluate a molecular dynamics simulation. We show that LoCoHD measurements provide unique information about protein structures that is distinct from, for example, those derived using the alignment-based RMSD metric, or the similarly distance matrix-based but alignment-free lDDT metric.

An improved multi-scale convolutional neural network with gated recurrent neural network model for protein secondary structure prediction

An improved multi-scale convolutional neural network with gated recurrent neural network model for protein secondary structure prediction

Heterologous expression and structure prediction of a xylanase identified from a compost metagenomic library

Xylanases are key biocatalysts in the degradation of the β‐1,4‐glycosidic linkages in the xylan backbone of hemicellulose. These enzymes are potentially applied in a wide range of bioprocessing industries under harsh conditions. Metagenomics has emerged as powerful tools for the bioprospection and discovery of interesting bioactive molecules from extreme ecosystems with unique features, such as high temperatures. In this study, an innovative combination of function-driven screening of a compost metagenomic library and automatic extraction of halo areas with in-house MATLAB functions resulted in the identification of a promising clone with xylanase activity (LP4). The LP4 clone proved to be an effective xylanase producer under submerged fermentation conditions. Sequence and phylogenetic analyses revealed that the xylanase, Xyl4, corresponded to an endo-1,4-β-xylanase belonging to glycosyl hydrolase family 10 (GH10). When xyl4 was expressed in Escherichia coli BL21(DE3), the enzyme activity increased about 2-fold compared to the LP4 clone. To get insight on the interaction of the enzyme with the substrate and establish possible strategies to improve its activity, the structure of Xyl4 was predicted, refined, and docked with xylohexaose. Our data unveiled, for the first time, the relevance of the amino acids Glu133 and Glu238 for catalysis, and a close inspection of the catalytic site suggested that the replacement of Phe316 by a bulkier Trp may improve Xyl4 activity. Our current findings contribute to enhancing the catalytic performance of Xyl4 towards industrial applications.Key points• A GH10 endo-1,4-β-xylanase (Xyl4) was isolated from a compost metagenomic library• MATLAB’s in-house functions were developed to identify the xylanase-producing clones• Computational analysis showed that Glu133 and Glu238 are crucial residues for catalysisGraphical abstract

Darobactin Substrate Engineering and Computation Show Radical Stability Governs Ether versus C-C Bond Formation.

The Gram-negative selective antibiotic darobactin A has attracted interest owing to its intriguing fused bicyclic structure and unique targeting of the outer membrane protein BamA. Darobactin, a ribosomally synthesized and post-translationally modified peptide (RiPP), is produced by a radical S-adenosyl methionine (rSAM)-dependent enzyme (DarE) and contains one ether and one C-C cross-link. Herein, we analyze the substrate tolerance of DarE and describe an underlying catalytic principle of the enzyme. These efforts produced 51 enzymatically modified darobactin variants, revealing that DarE can install the ether and C-C cross-links independently and in different locations on the substrate. Notable variants with fused bicyclic structures were characterized, including darobactin W3Y, with a non-Trp residue at the twice-modified central position, and darobactin K5F, which displays a fused diether ring pattern. While lacking antibiotic activity, quantum mechanical modeling of darobactins W3Y and K5F aided in the elucidation of the requisite features for high-affinity BamA engagement. We also provide experimental evidence for β-oxo modification, which adds support for a proposed DarE mechanism. Based on these results, ether and C-C cross-link formation was investigated computationally, and it was determined that more stable and longer-lived aromatic Cβ radicals correlated with ether formation. Further, molecular docking and transition state structures based on high-level quantum mechanical calculations support the different indole connectivity observed for ether (Trp-C7) and C-C (Trp-C6) cross-links. Finally, mutational analysis and protein structural predictions identified substrate residues that govern engagement to DarE. Our work informs on darobactin scaffold engineering and further unveils the underlying principles of rSAM catalysis.

DeepPPAPredMut: deep ensemble method for predicting the binding affinity change in protein-protein complexes upon mutation.

Protein-protein interactions (PPIs) underpin many cellular processes and their disruption due to mutations can lead to diseases. With the evolution of protein structure prediction methods like AlphaFold2 and the availability of extensive experimental affinity data, there is a pressing need for updated computational tools that can efficiently predict changes in binding affinity caused by mutations in protein-protein complexes. We developed a deep ensemble model that leverages protein sequences, predicted structure-based features, and protein functional classes to accurately predict the change in binding affinity due to mutations. The model achieved a correlation of 0.97 and a mean absolute error (MAE) of 0.35 kcal/mol on the training dataset, and maintained robust performance on the test set with a correlation of 0.72 and a MAE of 0.83 kcal/mol. Further validation using Leave-One-Out Complex (LOOC) cross-validation exhibited a correlation of 0.83 and a MAE of 0.51 kcal/mol, indicating consistent performance. https://web.iitm.ac.in/bioinfo2/DeepPPAPredMut/index.html. Supplementary data are available at Bioinformatics online.

Chainsaw: protein domain segmentation with fully convolutional neural networks.

Protein domains are fundamental units of protein structure and play a pivotal role in understanding folding, function, evolution, and design. The advent of accurate structure prediction techniques has resulted in an influx of new structural data, making the partitioning of these structures into domains essential for inferring evolutionary relationships and functional classification. This paper presents Chainsaw, a supervised learning approach to domain parsing that achieves accuracy that surpasses current state-of-the-art methods. Chainsaw uses a fully convolutional neural network which is trained to predict the probability that each pair of residues is in the same domain. Domain predictions are then derived from these pairwise predictions using an algorithm that searches for the most likely assignment of residues to domains given the set of pairwise co-membership probabilities. Chainsaw matches CATH domain annotations in 78% of protein domains versus 72% for the next closest method. When predicting on AlphaFold models, expert human evaluators were twice as likely to prefer Chainsaw's predictions versus the next best method. github.com/JudeWells/chainsaw.

Accurate structure prediction of biomolecular interactions with AlphaFold 3.

The introduction of AlphaFold 21 has spurred a revolution in modelling the structure of proteins and their interactions, enabling a huge range of applications in protein modelling and design2-6. In this paper, we describe our AlphaFold 3 model with a substantially updated diffusion-based architecture, which is capable of joint structure prediction of complexes including proteins, nucleic acids, small molecules, ions, and modified residues. The new AlphaFold model demonstrates significantly improved accuracy over many previous specialised tools: far greater accuracy on protein-ligand interactions than state of the art docking tools, much higher accuracy on protein-nucleic acid interactions than nucleic-acid-specific predictors, and significantly higher antibody-antigen prediction accuracy than AlphaFold-Multimer v2.37,8. Together these results show that high accuracy modelling across biomolecular space is possible within a single unified deep learning framework.