Improved prediction of protein-protein interactions using AlphaFold2

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 noncoding 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 intrinsically disordered regions (such as random coils), secondary structure elements (such as α helix and β strand), hydrophobicity, and molecular recognition features. In de novo proteins, specifically, we observed an 8% to 14% decay in random coils and intrinsically disordered region lengths and a 2.3% to 6.5% increase in structured elements, hydrophobicity, and molecular recognition features, 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 noncoding sequences, highlighting their quick transformation into active, protein complex-forming components within a remarkably short evolutionary timeframe.

MoLPC2: improved prediction of large protein complex structures and stoichiometry using Monte Carlo Tree Search and AlphaFold2

MOTIVATION

Today, the prediction of structures of large protein complexes solely from their sequence information requires prior knowledge of the stoichiometry of the complex. To address this challenge, we have enhanced the Monte Carlo Tree Search algorithms in MoLPC to enable the assembly of protein complexes while simultaneously predicting their stoichiometry.

RESULTS

In MoLPC2, we have improved the predictions by allowing sampling alternative AlphaFold predictions. Using MoLPC2, we accurately predicted the structures of 50 out of 175 nonredundant protein complexes (TM-score ≥ 0.8) without knowing the stoichiometry. MoLPC2 provides new opportunities for predicting protein complex structures without stoichiometry information.

AVAILABILITY AND IMPLEMENTATION

MoLPC2 is freely available at https://github.com/hychim/molpc2. A notebook is also available from the repository for easy use.

Apprehensions and emerging solutions in ML-based protein structure prediction

The three-dimensional structure of proteins determines their function in vital biological processes. Thus, when the structure is known, the molecular mechanism of protein function can be understood in more detail and obtained information utilized in biotechnological, diagnostics, and therapeutic applications. Over the past five years, machine learning (ML)-based modeling has pushed protein structure prediction to the next level with AlphaFold in the front line, predicting the structure for hundreds of millions of proteins. Further advances recently report promising ML-based approaches for solving remaining challenges by incorporating functionally important metals, co-factors, post-translational modifications, structural dynamics, and interdomain and multimer interactions in the structure prediction process.

Identification of type VI secretion system effector-immunity pairs using structural bioinformatics

The type VI secretion system (T6SS) is an important mediator of microbe-microbe and microbe-host interactions. Gram-negative bacteria use the T6SS to inject T6SS effectors (T6Es), which are usually proteins with toxic activity, into neighboring cells. Antibacterial effectors have cognate immunity proteins that neutralize self-intoxication. Here, we applied novel structural bioinformatic tools to perform systematic discovery and functional annotation of T6Es and their cognate immunity proteins from a dataset of 17,920 T6SS-encoding bacterial genomes. Using structural clustering, we identified 517 putative T6E families, outperforming sequence-based clustering. We developed a logistic regression model to reliably quantify protein-protein interaction of new T6E-immunity pairs, yielding candidate immunity proteins for 231 out of the 517 T6E families. We used sensitive structure-based annotation which yielded functional annotations for 51% of the T6E families, again outperforming sequence-based annotation. Next, we validated four novel T6E-immunity pairs using basic experiments in E. coli. In particular, we showed that the Pfam domain DUF3289 is a homolog of Colicin M and that DUF943 acts as its cognate immunity protein. Furthermore, we discovered a novel T6E that is a structural homolog of SleB, a lytic transglycosylase, and identified a specific glutamate that acts as its putative catalytic residue. Overall, this study applies novel structural bioinformatic tools to T6E-immunity pair discovery, and provides an extensive database of annotated T6E-immunity pairs.

Overview of AlphaFold2 and breakthroughs in overcoming its limitations

Predicting three-dimensional (3D) protein structures has been challenging for decades. The emergence of AlphaFold2 (AF2), a deep learning-based machine learning method developed by DeepMind, became a game changer in the protein folding community. AF2 can predict a protein's three-dimensional structure with high confidence based on its amino acid sequence. Accurate prediction of protein structures can dramatically accelerate our understanding of biological mechanisms and provide a solid foundation for reliable drug design. Although AF2 breaks through the barriers in predicting protein structures, many rooms remain to be further studied. This review provides a brief historical overview of the development of protein structure prediction, covering template-based, template-free, and machine learning-based methods. In addition to reviewing the potential benefits (Pros) and considerations (Cons) of using AF2, this review summarizes the diverse applications, including protein structure predictions, dynamic changes, point mutation, integration of language model and experimental data, protein complex, and protein-peptide interaction. It underscores recent advancements in efficiency, reliability, and broad application of AF2. This comprehensive review offers valuable insights into the applications of AF2 and AF2-inspired AI methods in structural biology and its potential for clinically significant drug target discovery.

Structural and functional insights into the C-terminal signal domain of the Bacteroidetes type-IX secretion system

Gram-negative bacteria from the Bacteroidota phylum possess a type-IX secretion system (T9SS) for protein secretion, which requires cargoes to have a C-terminal domain (CTD). Structurally analysed CTDs are from Porphyromonas gingivalis proteins RgpB, HBP35, PorU and PorZ, which share a compact immunoglobulin-like antiparallel 3+4 β-sandwich (β1-β7). This architecture is essential as a P. gingivalis strain with a single-point mutant of RgpB disrupting the interaction of the CTD with its preceding domain prevented secretion of the protein. Next, we identified the C-terminus ('motif C-t.') and the loop connecting strands β3 and β4 ('motif Lβ3β4') as conserved. We generated two strains with insertion and replacement mutants of PorU, as well as three strains with ablation and point mutants of RgpB, which revealed both motifs to be relevant for T9SS function. Furthermore, we determined the crystal structure of the CTD of mirolase, a cargo of the Tannerella forsythia T9SS, which shares the same general topology as in Porphyromonas CTDs. However, motif Lβ3β4 was not conserved. Consistently, P. gingivalis could not properly secrete a chimaeric protein with the CTD of peptidylarginine deiminase replaced with this foreign CTD. Thus, the incompatibility of the CTDs between these species prevents potential interference between their T9SSs.

Assessing the functional impact of protein binding site definition

Many biomedical applications, such as classification of binding specificities or bioengineering, depend on the accurate definition of protein binding interfaces. Depending on the choice of method used, substantially different sets of residues can be classified as belonging to the interface of a protein. A typical approach used to verify these definitions is to mutate residues and measure the impact of these changes on binding. Besides the lack of exhaustive data, this approach also suffers from the fundamental problem that a mutation introduces an unknown amount of alteration into an interface, which potentially alters the binding characteristics of the interface. In this study we explore the impact of alternative binding site definitions on the ability of a protein to recognize its cognate ligand using a pharmacophore approach, which does not affect the interface. The study also shows that methods for protein binding interface predictions should perform above approximately F-score = 0.7 accuracy level to capture the biological function of a protein.

Review and Comparative Analysis of Methods and Advancements in Predicting Protein Complex Structure

Protein complexes perform diverse biological functions, and obtaining their three-dimensional structure is critical to understanding and grasping their functions. In many cases, it's not just two proteins interacting to form a dimer; instead, multiple proteins interact to form a multimer. Experimentally resolving protein complex structures can be quite challenging. Recently, there have been efforts and methods that build upon prior predictions of dimer structures to attempt to predict multimer structures. However, in comparison to monomeric protein structure prediction, the accuracy of protein complex structure prediction remains relatively low. This paper provides an overview of recent advancements in efficient computational models for predicting protein complex structures. We introduce protein-protein docking methods in detail and summarize their main ideas, applicable modes, and related information. To enhance prediction accuracy, other critical protein-related information is also integrated, such as predicting interchain residue contact, utilizing experimental data like cryo-EM experiments, and considering protein interactions and non-interactions. In addition, we comprehensively review computational approaches for end-to-end prediction of protein complex structures based on artificial intelligence (AI) technology and describe commonly used datasets and representative evaluation metrics in protein complexes. Finally, we analyze the formidable challenges faced in current protein complex structure prediction tasks, including the structure prediction of heteromeric complex, disordered regions in complex, antibody-antigen complex, and RNA-related complex, as well as the evaluation metrics for complex assessment. We hope that this work will provide comprehensive knowledge of complex structure predictions to contribute to future advanced predictions.

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. Here we describe our AlphaFold 3 model with a substantially updated diffusion-based architecture that is capable of predicting the joint structure of complexes including proteins, nucleic acids, small molecules, ions and modified residues. The new AlphaFold model demonstrates substantially improved accuracy over many previous specialized tools: far greater accuracy for protein-ligand interactions compared with state-of-the-art docking tools, much higher accuracy for protein-nucleic acid interactions compared with nucleic-acid-specific predictors and substantially higher antibody-antigen prediction accuracy compared with AlphaFold-Multimer v.2.37,8. Together, these results show that high-accuracy modelling across biomolecular space is possible within a single unified deep-learning framework.

Decoding the Arsenal: Protist Effectors and Their Impact on Photosynthetic Hosts

Interactions between various microbial pathogens including viruses, bacteria, fungi, oomycetes, and their plant hosts have traditionally been the focus of phytopathology. In recent years, a significant and growing interest in the study of eukaryotic microorganisms not classified among fungi or oomycetes has emerged. Many of these protists establish complex interactions with photosynthetic hosts, and understanding these interactions is crucial in understanding the dynamics of these parasites within traditional and emerging types of farming, including marine aquaculture. Many phytopathogenic protists are biotrophs with complex polyphasic life cycles, which makes them difficult or impossible to culture, a fact reflected in a wide gap in the availability of comprehensive genomic data when compared to fungal and oomycete plant pathogens. Furthermore, our ability to use available genomic resources for these protists is limited by the broad taxonomic distance that these organisms span, which makes comparisons with other genomic datasets difficult. The current rapid progress in genomics and computational tools for the prediction of protein functions and interactions is revolutionizing the landscape in plant pathology. This is also opening novel possibilities, specifically for a deeper understanding of protist effectors. Tools like AlphaFold2 enable structure-based function prediction of effector candidates with divergent protein sequences. In turn, this allows us to ask better biological questions and, coupled with innovative experimental strategies, will lead into a new era of effector research, especially for protists, to expand our knowledge on these elusive pathogens and their interactions with photosynthetic hosts. [Formula: see text] Copyright © 2024 The Author(s). This is an open access article distributed under the CC BY-NC-ND 4.0 International license.

From GPUs to AI and quantum: three waves of acceleration in bioinformatics

The enormous growth in the amount of data generated by the life sciences is continuously shifting the field from model-driven science towards data-driven science. The need for efficient processing has led to the adoption of massively parallel accelerators such as graphics processing units (GPUs). Consequently, the development of bioinformatics methods nowadays often heavily depends on the effective use of these powerful technologies. Furthermore, progress in computational techniques and architectures continues to be highly dynamic, involving novel deep neural network models and artificial intelligence (AI) accelerators, and potentially quantum processing units in the future. These are expected to be disruptive for the life sciences as a whole and for drug discovery in particular. Here, we identify three waves of acceleration and their applications in a bioinformatics context: (i) GPU computing, (ii) AI and (iii) next-generation quantum computers.

Defying gravity: WEEP promotes negative gravitropism in peach trees by establishing asymmetric auxin gradients

Trees with weeping shoot architectures are valued for their beauty and are a resource for understanding how plants regulate posture control. The peach (Prunus persica) weeping phenotype, which has elliptical downward arching branches, is caused by a homozygous mutation in the WEEP gene. Little is known about the function of WEEP despite its high conservation throughout Plantae. Here, we present the results of anatomical, biochemical, biomechanical, physiological, and molecular experiments that provide insight into WEEP function. Our data suggest that weeping peach trees do not have defects in branch structure. Rather, transcriptomes from the adaxial (upper) and abaxial (lower) sides of standard and weeping branch shoot tips revealed flipped expression patterns for genes associated with early auxin response, tissue patterning, cell elongation, and tension wood development. This suggests that WEEP promotes polar auxin transport toward the lower side during shoot gravitropic response, leading to cell elongation and tension wood development. In addition, weeping peach trees exhibited steeper root systems and faster lateral root gravitropic response. This suggests that WEEP moderates root gravitropism and is essential to establishing the set-point angle of lateral roots from the gravity vector. Additionally, size exclusion chromatography indicated that WEEP proteins self-oligomerize, like other proteins with sterile alpha motif domains. Collectively, our results from weeping peach provide insight into polar auxin transport mechanisms associated with gravitropism and lateral shoot and root orientation.

GTExome: Modeling commonly expressed missense mutations in the human genome

A web application, GTExome, is described that quickly identifies, classifies, and models missense mutations in commonly expressed human proteins. GTExome can be used to categorize genomic mutation data with tissue specific expression data from the Genotype-Tissue Expression (GTEx) project. Commonly expressed missense mutations in proteins from a wide range of tissue types can be selected and assessed for modeling suitability. Information about the consequences of each mutation is provided to the user including if disulfide bonds, hydrogen bonds, or salt bridges are broken, buried prolines introduced, buried charges are created or lost, charge is swapped, a buried glycine is replaced, or if the residue that would be removed is a proline in the cis configuration. Also, if the mutation site is in a binding pocket the number of pockets and their volumes are reported. The user can assess this information and then select from available experimental or computationally predicted structures of native proteins to create, visualize, and download a model of the mutated protein using Fast and Accurate Side-chain Protein Repacking (FASPR). For AlphaFold modeled proteins, confidence scores for native proteins are provided. Using this tool, we explored a set of 9,666 common missense mutations from a variety of tissues from GTEx and show that most mutations can be modeled using this tool to facilitate studies of protein-protein and protein-drug interactions. The open-source tool is freely available at https://pharmacogenomics.clas.ucdenver.edu/gtexome/.

Efficient Refinement of Complex Structures of Flexible Histone Peptides Using Post-Docking Molecular Dynamics Protocols

Histones are keys to many epigenetic events and their complexes have therapeutic and diagnostic importance. The determination of the structures of histone complexes is fundamental in the design of new drugs. Computational molecular docking is widely used for the prediction of target-ligand complexes. Large, linear peptides like the tail regions of histones are challenging ligands for docking due to their large conformational flexibility, extensive hydration, and weak interactions with the shallow binding pockets of their reader proteins. Thus, fast docking methods often fail to produce complex structures of such peptide ligands at a level appropriate for drug design. To address this challenge, and improve the structural quality of the docked complexes, post-docking refinement has been applied using various molecular dynamics (MD) approaches. However, a final consensus has not been reached on the desired MD refinement protocol. In this present study, MD refinement strategies were systematically explored on a set of problematic complexes of histone peptide ligands with relatively large errors in their docked geometries. Six protocols were compared that differ in their MD simulation parameters. In all cases, pre-MD hydration of the complex interface regions was applied to avoid the unwanted presence of empty cavities. The best-performing protocol achieved a median of 32% improvement over the docked structures in terms of the change in root mean squared deviations from the experimental references. The influence of structural factors and explicit hydration on the performance of post-docking MD refinements are also discussed to help with their implementation in future methods and applications.

Predicted mechanistic impacts of human protein missense variants

Genome sequencing efforts have led to the discovery of tens of millions of protein missense variants found in the human population with the majority of these having no annotated role and some likely contributing to trait variation and disease. Sequence-based artificial intelligence approaches have become highly accurate at predicting variants that are detrimental to the function of proteins but they do not inform on mechanisms of disruption. Here we combined sequence and structure-based methods to perform proteome-wide prediction of deleterious variants with information on their impact on protein stability, protein-protein interactions and small-molecule binding pockets. AlphaFold2 structures were used to predict approximately 100,000 small-molecule binding pockets and stability changes for over 200 million variants. To inform on protein-protein interfaces we used AlphaFold2 to predict structures for nearly 500,000 protein complexes. We illustrate the value of mechanism-aware variant effect predictions to study the relation between protein stability and abundance and the structural properties of interfaces underlying trans protein quantitative trait loci (pQTLs). We characterised the distribution of mechanistic impacts of protein variants found in patients and experimentally studied example disease linked variants in FGFR1.

An Ensemble Classifiers for Improved Prediction of Native-Non-Native Protein-Protein Interaction

In this study, we present an innovative approach to improve the prediction of protein-protein interactions (PPIs) through the utilization of an ensemble classifier, specifically focusing on distinguishing between native and non-native interactions. Leveraging the strengths of various base models, including random forest, gradient boosting, extreme gradient boosting, and light gradient boosting, our ensemble classifier integrates these diverse predictions using a logistic regression meta-classifier. Our model was evaluated using a comprehensive dataset generated from molecular dynamics simulations. While the gains in AUC and other metrics might seem modest, they contribute to a model that is more robust, consistent, and adaptable. To assess the effectiveness of various approaches, we compared the performance of logistic regression to four baseline models. Our results indicate that logistic regression consistently underperforms across all evaluated metrics. This suggests that it may not be well-suited to capture the complex relationships within this dataset. Tree-based models, on the other hand, appear to be more effective for problems involving molecular dynamics simulations. Extreme gradient boosting (XGBoost) and light gradient boosting (LightGBM) are optimized for performance and speed, handling datasets effectively and incorporating regularizations to avoid over-fitting. Our findings indicate that the ensemble method enhances the predictive capability of PPIs, offering a promising tool for computational biology and drug discovery by accurately identifying potential interaction sites and facilitating the understanding of complex protein functions within biological systems.

GTE: a graph learning framework for prediction of T-cell receptors and epitopes binding specificity

The interaction between T-cell receptors (TCRs) and peptides (epitopes) presented by major histocompatibility complex molecules (MHC) is fundamental to the immune response. Accurate prediction of TCR-epitope interactions is crucial for advancing the understanding of various diseases and their prevention and treatment. Existing methods primarily rely on sequence-based approaches, overlooking the inherent topology structure of TCR-epitope interaction networks. In this study, we present $GTE$, a novel heterogeneous Graph neural network model based on inductive learning to capture the topological structure between TCRs and Epitopes. Furthermore, we address the challenge of constructing negative samples within the graph by proposing a dynamic edge update strategy, enhancing model learning with the nonbinding TCR-epitope pairs. Additionally, to overcome data imbalance, we adapt the Deep AUC Maximization strategy to the graph domain. Extensive experiments are conducted on four public datasets to demonstrate the superiority of exploring underlying topological structures in predicting TCR-epitope interactions, illustrating the benefits of delving into complex molecular networks. The implementation code and data are available at https://github.com/uta-smile/GTE.

A new mRNA structure prediction based approach to identifying improved signal peptides for bone morphogenetic protein 2

BACKGROUND

Signal peptide (SP) engineering has proven able to improve production of many proteins yet is a laborious process that still relies on trial and error. mRNA structure around the translational start site is important in translation initiation and has rarely been considered in this context, with recent improvements in in silico mRNA structure potentially rendering it a useful predictive tool for SP selection. Here we attempt to create a method to systematically screen candidate signal peptide sequences in silico based on both their nucleotide and amino acid sequences. Several recently released computational tools were used to predict signal peptide activity (SignalP), localization target (DeepLoc) and predicted mRNA structure (MXFold2). The method was tested with Bone Morphogenetic Protein 2 (BMP2), an osteogenic growth factor used clinically for bone regeneration. It was hoped more effective BMP2 SPs could improve BMP2-based gene therapies and reduce the cost of recombinant BMP2 production.

RESULTS

Amino acid sequence analysis indicated 2,611 SPs from the TGF-β superfamily were predicted to function when attached to BMP2. mRNA structure prediction indicated structures at the translational start site were likely highly variable. The five sequences with the most accessible translational start sites, a codon optimized BMP2 SP variant and the well-established hIL2 SP sequence were taken forward to in vitro testing. The top five candidates showed non-significant improvements in BMP2 secretion in HEK293T cells. All showed reductions in secretion versus the native sequence in C2C12 cells, with several showing large and significant decreases. None of the tested sequences were able to increase alkaline phosphatase activity above background in C2C12s. The codon optimized control sequence and hIL2 SP showed reasonable activity in HEK293T but very poor activity in C2C12.

CONCLUSIONS

These results support the use of peptide sequence based in silico tools for basic predictions around signal peptide activity in a synthetic biology context. However, mRNA structure prediction requires improvement before it can produce reliable predictions for this application. The poor activity of the codon optimized BMP2 SP variant in C2C12 emphasizes the importance of codon choice, mRNA structure, and cellular context for SP activity.

Genotype and phenotype correlation of PHACTR1-related neurological disorders

BACKGROUND

PHACTR1 (phosphatase and actin regulators) plays a key role in cortical migration and synaptic activity by binding and regulating G-actin and PPP1CA. This study aimed to expand the genotype and phenotype of patients with de novo variants in PHACTR1 and analyse the impact of variants on protein-protein interaction.

METHODS

We identified seven patients with PHACTR1 variants by trio-based whole-exome sequencing. Additional two subjects were ascertained from two centres through GeneMatcher. The genotype-phenotype correlation was determined, and AlphaFold-Multimer was used to predict protein-protein interactions and interfaces.

RESULTS

Eight individuals carried missense variants and one had CNV in the PHACTR1. Infantile epileptic spasms syndrome (IESS) was the unifying phenotype in eight patients with missense variants of PHACTR1. They could present with other types of seizures and often exhibit drug-resistant epilepsy with a poor prognosis. One patient with CNV displayed a developmental encephalopathy phenotype. Using AlphaFold-Multimer, our findings indicate that PHACTR1 and G-actin-binding sequences overlap with PPP1CA at the RPEL3 domain, which suggests possible competition between PPP1CA and G-actin for binding to PHACTR1 through a similar polymerisation interface. In addition, patients carrying missense variants located at the PHACTR1-PPP1CA or PHACTR1-G-actin interfaces consistently exhibit the IESS phenotype. These missense variants are mostly concentrated in the overlapping sequence (RPEL3 domain).

CONCLUSIONS

Patients with variants in PHACTR1 can have a phenotype of developmental encephalopathy in addition to IESS. Moreover, our study confirmed that the variants affect the binding of PHACTR1 to G-actin or PPP1CA, resulting in neurological disorders in patients.

Semaphorin3C identified as mediator of neuroinflammation and microglia polarization after spinal cord injury

Excessive neuroinflammation after spinal cord injury (SCI) is a major hurdle during nerve repair. Although proinflammatory macrophage/microglia-mediated neuroinflammation plays important roles, the underlying mechanism that triggers neuroinflammation and aggravating factors remain unclear. The present study identified a proinflammatory role of semaphorin3C (SEMA3C) in immunoregulation after SCI. SEMA3C expression level peaked 7 days post-injury (dpi) and decreased by 14 dpi. In vivo and in vitro studies revealed that macrophages/microglia expressed SEMA3C in the local microenvironment, which induced neuroinflammation and conversion of proinflammatory macrophage/microglia. Mechanistic experiments revealed that RAGE/NF-κB was downstream target of SEMA3C. Inhibiting SEMA3C-mediated RAGE signaling considerably suppressed proinflammatory cytokine production, reversed polarization of macrophages/microglia shortly after SCI. In addition, inhibition of SEMA3C-mediated RAGE signaling suggested that the SEMA3C/RAGE axis is a feasible target to preserve axons from neuroinflammation. Taken together, our study provides the first experimental evidence of an immunoregulatory role for SEMA3C in SCI via an autocrine mechanism.

Beyond the Triple Helix: Exploration of the Hierarchical Assembly Space of Collagen-like Peptides

The de novo design of self-assembling peptides has garnered significant attention in scientific research. While alpha-helical assemblies have been extensively studied, exploration of polyproline type II (PPII) helices, such as those found in collagen, remains relatively limited. In this study, we focused on understanding the sequence-structure relationship in hierarchical assemblies of collagen-like peptides, using defense collagen SP-A as a model. By dissecting the sequence derived from SP-A and synthesizing short collagen-like peptides, we successfully constructed a discrete bundle of hollow triple helices. Mutation studies pinpointed amino acid sequences, including hydrophobic and charged residues that are critical for oligomer formation. These insights guided the de novo design of collagen-like peptides, resulting in the formation of diverse quaternary structures, including discrete and heterogenous bundled oligomers, 2D nanosheets, and pH-responsive nanoribbons. Our study represents a significant advancement in the understanding and harnessing of collagen higher-order assemblies beyond the triple helix.

A Survey of Deep Learning Methods for Estimating the Accuracy of Protein Quaternary Structure Models

The quality prediction of quaternary structure models of a protein complex, in the absence of its true structure, is known as the Estimation of Model Accuracy (EMA). EMA is useful for ranking predicted protein complex structures and using them appropriately in biomedical research, such as protein-protein interaction studies, protein design, and drug discovery. With the advent of more accurate protein complex (multimer) prediction tools, such as AlphaFold2-Multimer and ESMFold, the estimation of the accuracy of protein complex structures has attracted increasing attention. Many deep learning methods have been developed to tackle this problem; however, there is a noticeable absence of a comprehensive overview of these methods to facilitate future development. Addressing this gap, we present a review of deep learning EMA methods for protein complex structures developed in the past several years, analyzing their methodologies, data and feature construction. We also provide a prospective summary of some potential new developments for further improving the accuracy of the EMA methods.

Protein features fusion using attributed network embedding for predicting protein-protein interaction

BACKGROUND

Protein-protein interactions (PPIs) hold significant importance in biology, with precise PPI prediction as a pivotal factor in comprehending cellular processes and facilitating drug design. However, experimental determination of PPIs is laborious, time-consuming, and often constrained by technical limitations.

METHODS

We introduce a new node representation method based on initial information fusion, called FFANE, which amalgamates PPI networks and protein sequence data to enhance the precision of PPIs' prediction. A Gaussian kernel similarity matrix is initially established by leveraging protein structural resemblances. Concurrently, protein sequence similarities are gauged using the Levenshtein distance, enabling the capture of diverse protein attributes. Subsequently, to construct an initial information matrix, these two feature matrices are merged by employing weighted fusion to achieve an organic amalgamation of structural and sequence details. To gain a more profound understanding of the amalgamated features, a Stacked Autoencoder (SAE) is employed for encoding learning, thereby yielding more representative feature representations. Ultimately, classification models are trained to predict PPIs by using the well-learned fusion feature.

RESULTS

When employing 5-fold cross-validation experiments on SVM, our proposed method achieved average accuracies of 94.28%, 97.69%, and 84.05% in terms of Saccharomyces cerevisiae, Homo sapiens, and Helicobacter pylori datasets, respectively.

CONCLUSION

Experimental findings across various authentic datasets validate the efficacy and superiority of this fusion feature representation approach, underscoring its potential value in bioinformatics.

Evaluation of AlphaFold2 Structures for Hit Identification across Multiple Scenarios

The introduction of AlphaFold2 (AF2) has sparked significant enthusiasm and generated extensive discussion within the scientific community, particularly among drug discovery researchers. Although previous studies have addressed the performance of AF2 structures in virtual screening (VS), a more comprehensive investigation is still necessary considering the paramount importance of structural accuracy in drug design. In this study, we evaluate the performance of AF2 structures in VS across three common drug discovery scenarios: targets with holo, apo, and AF2 structures; targets with only apo and AF2 structures; and targets exclusively with AF2 structures. We utilized both the traditional physics-based Glide and the deep-learning-based scoring function RTMscore to rank the compounds in the DUD-E, DEKOIS 2.0, and DECOY data sets. The results demonstrate that, overall, the performance of VS on AF2 structures is comparable to that on apo structures but notably inferior to that on holo structures across diverse scenarios. Moreover, when a target has solely AF2 structure, selecting the holo structure of the target from different subtypes within the same protein family produces comparable results with the AF2 structure for VS on the data set of the AF2 structures, and significantly better results than the AF2 structures on its own data set. This indicates that utilizing AF2 structures for docking-based VS may not yield most satisfactory outcomes, even when solely AF2 structures are available. Moreover, we rule out the possibility that the variations in VS performance between the binding pockets of AF2 and holo structures arise from the differences in their biological assembly composition.

A Hadal Streptomyces-Derived Echinocandin Acylase Discovered through the Prioritization of Protein Families

Naturally occurring echinocandin B and FR901379 are potent antifungal lipopeptides featuring a cyclic hexapeptide nucleus and a fatty acid side chain. They are the parent compounds of echinocandin drugs for the treatment of severe fungal infections caused by the Candida and Aspergilla species. To minimize hemolytic toxicity, the native fatty acid side chains in these drug molecules are replaced with designer acyl side chains. The deacylation of the N-acyl side chain is, therefore, a crucial step for the development and manufacturing of echinocandin-type antibiotics. Echinocandin E (ECE) is a novel echinocandin congener with enhanced stability generated via the engineering of the biosynthetic machinery of echinocandin B (ECB). In the present study, we report the discovery of the first echinocandin E acylase (ECEA) using the enzyme similarity tool (EST) for enzymatic function mining across protein families. ECEA is derived from Streptomyces sp. SY1965 isolated from a sediment collected from the Mariana Trench. It was cloned and heterologously expressed in S. lividans TK24. The resultant TKecea66 strain showed efficient cleavage activity of the acyl side chain of ECE, showing promising applications in the development of novel echinocandin-type therapeutics. Our results also provide a showcase for harnessing the essentially untapped biodiversity from the hadal ecosystems for the discovery of functional molecules.

FuncPhos-STR: An integrated deep neural network for functional phosphosite prediction based on AlphaFold protein structure and dynamics

Phosphorylation modifications play important regulatory roles in most biological processes. However, the functional assignment for the vast majority of the identified phosphosites remains a major challenge. Here, we provide a deep learning framework named FuncPhos-STR as an online resource, for functional prediction and structural visualization of human proteome-level phosphosites. Based on our reported FuncPhos-SEQ framework, which was built by integrating phosphosite sequence evolution and protein-protein interaction (PPI) information, FuncPhos-STR was developed by further integrating the structural and dynamics information on AlphaFold protein structures. The characterized structural topology and dynamics features underlying functional phosphosites emphasized their molecular mechanism for regulating protein functions. By integrating the structural and dynamics, sequence evolutionary, and PPI network features from protein different dimensions, FuncPhos-STR has advantage over other reported models, with the best AUC value of 0.855. Using FuncPhos-STR, the phosphosites inside the pocket regions are accessible to higher functional scores, theoretically supporting their potential regulatory mechanism. Overall, FuncPhos-STR would accelerate the functional identification of huge unexplored phosphosites, and facilitate the elucidation of their allosteric regulation mechanisms. The web server of FuncPhos-STR is freely available at http://funcptm.jysw.suda.edu.cn/str.

The structural landscape and diversity of Pyricularia oryzae MAX effectors revisited

Magnaporthe AVRs and ToxB-like (MAX) effectors constitute a family of secreted virulence proteins in the fungus Pyricularia oryzae (syn. Magnaporthe oryzae), which causes blast disease on numerous cereals and grasses. In spite of high sequence divergence, MAX effectors share a common fold characterized by a ß-sandwich core stabilized by a conserved disulfide bond. In this study, we investigated the structural landscape and diversity within the MAX effector repertoire of P. oryzae. Combining experimental protein structure determination and in silico structure modeling we validated the presence of the conserved MAX effector core domain in 77 out of 94 groups of orthologs (OG) identified in a previous population genomic study. Four novel MAX effector structures determined by NMR were in remarkably good agreement with AlphaFold2 (AF2) predictions. Based on the comparison of the AF2-generated 3D models we propose a classification of the MAX effectors superfamily in 20 structural groups that vary in the canonical MAX fold, disulfide bond patterns, and additional secondary structures in N- and C-terminal extensions. About one-third of the MAX family members remain singletons, without strong structural relationship to other MAX effectors. Analysis of the surface properties of the AF2 MAX models also highlights the high variability within the MAX family at the structural level, potentially reflecting the wide diversity of their virulence functions and host targets.

PRRX1-TOP2A interaction is a malignancy-promoting factor in human malignant peripheral nerve sheath tumours

BACKGROUND

Paired related-homeobox 1 (PRRX1) is a transcription factor in the regulation of developmental morphogenetic processes. There is growing evidence that PRRX1 is highly expressed in certain cancers and is critically involved in human survival prognosis. However, the molecular mechanism of PRRX1 in cancer malignancy remains to be elucidated.

METHODS

PRRX1 expression in human Malignant peripheral nerve sheath tumours (MPNSTs) samples was detected immunohistochemically to evaluate survival prognosis. MPNST models with PRRX1 gene knockdown or overexpression were constructed in vitro and the phenotype of MPNST cells was evaluated. Bioinformatics analysis combined with co-immunoprecipitation, mass spectrometry, RNA-seq and structural prediction were used to identify proteins interacting with PRRX1.

RESULTS

High expression of PRRX1 was associated with a poor prognosis for MPNST. PRRX1 knockdown suppressed the tumorigenic potential. PRRX1 overexpressed in MPNSTs directly interacts with topoisomerase 2 A (TOP2A) to cooperatively promote epithelial-mesenchymal transition and increase expression of tumour malignancy-related gene sets including mTORC1, KRAS and SRC signalling pathways. Etoposide, a TOP2A inhibitor used in the treatment of MPNST, may exhibit one of its anticancer effects by inhibiting the PRRX1-TOP2A interaction.

CONCLUSION

Targeting the PRRX1-TOP2A interaction in malignant tumours with high PRRX1 expression might provide a novel tumour-selective therapeutic strategy.

PPIscreenML: Structure-based screening for protein-protein interactions using AlphaFold

Protein-protein interactions underlie nearly all cellular processes. With the advent of protein structure prediction methods such as AlphaFold2 (AF2), models of specific protein pairs can be built extremely accurately in most cases. However, determining the relevance of a given protein pair remains an open question. It is presently unclear how to use best structure-based tools to infer whether a pair of candidate proteins indeed interact with one another: ideally, one might even use such information to screen amongst candidate pairings to build up protein interaction networks. Whereas methods for evaluating quality of modeled protein complexes have been co-opted for determining which pairings interact (e.g., pDockQ and iPTM), there have been no rigorously benchmarked methods for this task. Here we introduce PPIscreenML, a classification model trained to distinguish AF2 models of interacting protein pairs from AF2 models of compelling decoy pairings. We find that PPIscreenML out-performs methods such as pDockQ and iPTM for this task, and further that PPIscreenML exhibits impressive performance when identifying which ligand/receptor pairings engage one another across the structurally conserved tumor necrosis factor superfamily (TNFSF). Analysis of benchmark results using complexes not seen in PPIscreenML development strongly suggest that the model generalizes beyond training data, making it broadly applicable for identifying new protein complexes based on structural models built with AF2.

Deep learning insights into the architecture of the mammalian egg-sperm fusion synapse

A crucial event in sexual reproduction is when haploid sperm and egg fuse to form a new diploid organism at fertilization. In mammals, direct interaction between egg JUNO and sperm IZUMO1 mediates gamete membrane adhesion, yet their role in fusion remains enigmatic. We used AlphaFold to predict the structure of other extracellular proteins essential for fertilization to determine if they could form a complex that may mediate fusion. We first identified TMEM81, whose gene is expressed by mouse and human spermatids, as a protein having structural homologies with both IZUMO1 and another sperm molecule essential for gamete fusion, SPACA6. Using a set of proteins known to be important for fertilization and TMEM81, we then systematically searched for predicted binary interactions using an unguided approach and identified a pentameric complex involving sperm IZUMO1, SPACA6, TMEM81 and egg JUNO, CD9. This complex is structurally consistent with both the expected topology on opposing gamete membranes and the location of predicted N-glycans not modeled by AlphaFold-Multimer, suggesting that its components could organize into a synapse-like assembly at the point of fusion. Finally, the structural modeling approach described here could be more generally useful to gain insights into transient protein complexes difficult to detect experimentally.

NFATC2IP is a mediator of SUMO-dependent genome integrity

The post-translational modification of proteins by SUMO is crucial for cellular viability and mammalian development in part due to the contribution of SUMOylation to genome duplication and repair. To investigate the mechanisms underpinning the essential function of SUMO, we undertook a genome-scale CRISPR/Cas9 screen probing the response to SUMOylation inhibition. This effort identified 130 genes whose disruption reduces or enhances the toxicity of TAK-981, a clinical-stage inhibitor of the SUMO E1-activating enzyme. Among the strongest hits, we validated and characterized NFATC2IP, an evolutionarily conserved protein related to the fungal Esc2 and Rad60 proteins that harbors tandem SUMO-like domains. Cells lacking NFATC2IP are viable but are hypersensitive to SUMO E1 inhibition, likely due to the accumulation of mitotic chromosome bridges and micronuclei. NFATC2IP primarily acts in interphase and associates with nascent DNA, suggesting a role in the postreplicative resolution of replication or recombination intermediates. Mechanistically, NFATC2IP interacts with the SMC5/6 complex and UBC9, the SUMO E2, via its first and second SUMO-like domains, respectively. AlphaFold-Multimer modeling suggests that NFATC2IP positions and activates the UBC9-NSMCE2 complex, the SUMO E3 ligase associated with SMC5/SMC6. We conclude that NFATC2IP is a key mediator of SUMO-dependent genomic integrity that collaborates with the SMC5/6 complex.

Screening androgen receptor agonists of fish species using machine learning and molecular model in NORMAN water-relevant list

Androgen receptor (AR) agonists have strong endocrine disrupting effects in fish. Most studies mainly investigate AR binding capacity using human AR in vitro. However, there is still few methods to rapidly predict AR agonists in aquatic organisms. This study aimed to screen AR agonists of fish species using machine learning and molecular models in water-relevant list from NORMAN, a network of reference laboratories for monitoring contaminants of emerging concern in the environment. In this study, machine learning approaches (e.g., Deep Forest (DF)), Random Forests and artificial neural networks) were applied to predict AR agonists. Zebrafish, fathead minnow, mosquitofish, medaka fish and grass carp are all important aquatic model organisms widely used to evaluate the toxicity of new pollutants, and the molecular models of ARs from these five fish species were constructed to further screen AR agonists using AlphaFold2. The DF method showed the best performances with 0.99 accuracy, 0.97 sensitivity and 1 precision. The Asn705, Gln711, Arg752, and Thr877 residues in human AR and the corresponding sites in ARs from the five fish species were responsible for agonist binding. Overall, 245 substances were predicted as suspect AR agonists in the five fish species, including, certain glucocorticoids, cholesterol metabolites, and cardiovascular drugs in the NORMAN list. Using machine learning and molecular modeling hybrid methods rapidly and accurately screened AR agonists in fish species, and helping evaluate their ecological risk in fish populations.

Predictomes: A classifier-curated database of AlphaFold-modeled protein-protein interactions

Protein-protein interactions (PPIs) are ubiquitous in biology, yet a comprehensive structural characterization of the PPIs underlying biochemical processes is lacking. Although AlphaFold-Multimer (AF-M) has the potential to fill this knowledge gap, standard AF-M confidence metrics do not reliably separate relevant PPIs from an abundance of false positive predictions. To address this limitation, we used machine learning on well curated datasets to train a Structure Prediction and Omics informed Classifier called SPOC that shows excellent performance in separating true and false PPIs, including in proteome-wide screens. We applied SPOC to an all-by-all matrix of nearly 300 human genome maintenance proteins, generating ~40,000 predictions that can be viewed at predictomes.org, where users can also score their own predictions with SPOC. High confidence PPIs discovered using our approach suggest novel hypotheses in genome maintenance. Our results provide a framework for interpreting large scale AF-M screens and help lay the foundation for a proteome-wide structural interactome.

Computational Approaches to Predict Protein-Protein Interactions in Crowded Cellular Environments

Investigating protein-protein interactions is crucial for understanding cellular biological processes because proteins often function within molecular complexes rather than in isolation. While experimental and computational methods have provided valuable insights into these interactions, they often overlook a critical factor: the crowded cellular environment. This environment significantly impacts protein behavior, including structural stability, diffusion, and ultimately the nature of binding. In this review, we discuss theoretical and computational approaches that allow the modeling of biological systems to guide and complement experiments and can thus significantly advance the investigation, and possibly the predictions, of protein-protein interactions in the crowded environment of cell cytoplasm. We explore topics such as statistical mechanics for lattice simulations, hydrodynamic interactions, diffusion processes in high-viscosity environments, and several methods based on molecular dynamics simulations. By synergistically leveraging methods from biophysics and computational biology, we review the state of the art of computational methods to study the impact of molecular crowding on protein-protein interactions and discuss its potential revolutionizing effects on the characterization of the human interactome.

CD74 as a prognostic and M1 macrophage infiltration marker in a comprehensive pan-cancer analysis

CD74 is a type-II transmembrane glycoprotein that has been linked to tumorigenesis. However, this association was based only on phenotypic studies, and, to date, no in-depth mechanistic studies have been conducted. In this study, combined with a multi-omics study, CD74 levels were significantly upregulated in most cancers relative to normal tissues and were found to be predictive of prognosis. Elevated CD74 expression was associated with reduced levels of mismatch-repair genes and homologous repair gene signatures in over 10 tumor types. Multiple fluorescence staining and bulk, spatial, single-cell transcriptional analyses indicated its potential as a marker for M1 macrophage infiltration in pan-cancer. In addition, CD74 expression was higher in BRCA patients responsive to conventional chemotherapy and was able to predict the prognosis of these patients. Potential CD74-activating drugs (HNHA and BRD-K55186349) were identified through molecular docking to CD74. The findings indicate activation of CD74 may have potential in tumor immunotherapy.

The Origin and Evolution of Sex Peptide and Sex Peptide Receptor Interactions

Post-mating responses play a vital role in successful reproduction across diverse species. In fruit flies, sex peptide binds to the sex peptide receptor, triggering a series of post-mating responses. However, the origin of sex peptide receptor predates the emergence of sex peptide. The evolutionary origins of the interactions between sex peptide and sex peptide receptor and the mechanisms by which they interact remain enigmatic. In this study, we used ancestral sequence reconstruction, AlphaFold2 predictions, and molecular dynamics simulations to study sex peptide-sex peptide receptor interactions and their origination. Using AlphaFold2 and long-time molecular dynamics simulations, we predicted the structure and dynamics of sex peptide-sex peptide receptor interactions. We show that sex peptide potentially binds to the ancestral states of Diptera sex peptide receptor. Notably, we found that only a few amino acid changes in sex peptide receptor are sufficient for the formation of sex peptide-sex peptide receptor interactions. Ancestral sequence reconstruction and molecular dynamics simulations further reveal that sex peptide receptor interacts with sex peptide through residues that are mostly involved in the interaction interface of an ancestral ligand, myoinhibitory peptides. We propose a potential mechanism whereby sex peptide-sex peptide receptor interactions arise from the preexisting myoinhibitory peptides-sex peptide receptor interface as well as early chance events both inside and outside the preexisting interface that created novel sex peptide-specific sex peptide-sex peptide receptor interactions. Our findings provide new insights into the origin and evolution of sex peptide-sex peptide receptor interactions and their relationship with myoinhibitory peptides-sex peptide receptor interactions.

MFA-DTI: Drug-target interaction prediction based on multi-feature fusion adopted framework

The identification of drug-target interactions (DTI) is a valuable step in the drug discovery and repositioning process. However, traditional laboratory experiments are time-consuming and expensive. Computational methods have streamlined research to determine DTIs. The application of deep learning methods has significantly improved the prediction performance for DTIs. Modern deep learning methods can leverage multiple sources of information, including sequence data that contains biological structural information, and interaction data. While useful, these methods cannot be effectively applied to each type of information individually (e.g., chemical structure and interaction network) and do not take into account the specificity of DTI data such as low- or zero-interaction biological entities. To overcome these limitations, we propose a method called MFA-DTI (Multi-feature Fusion Adopted framework for DTI). MFA-DTI consists of three modules: an interaction graph learning module that processes the interaction network to generate interaction vectors, a chemical structure learning module that extracts features from the chemical structure, and a fusion module that combines these features for the final prediction. To validate the performance of MFA-DTI, we conducted experiments on six public datasets under different settings. The results indicate that the proposed method is highly effective in various settings and outperforms state-of-the-art methods.

Interfacial residues in protein-protein complexes are in the eyes of the beholder

Interactions between proteins are vital in almost all biological processes. The characterization of protein-protein interactions helps us understand the mechanistic basis of biological processes, thereby enabling the manipulation of proteins for biotechnological and clinical purposes. The interface residues of a protein-protein complex are assumed to have the following two properties: (a) they always interact with a residue of a partner protein, which forms the basis for distance-based interface residue identification methods, and (b) they are solvent-exposed in the isolated form of the protein and become buried in the complex form, which forms the basis for Accessible Surface Area (ASA)-based methods. The study interrogates this popular assumption by recognizing interface residues in protein-protein complexes through these two methods. The results show that a few residues are identified uniquely by each method, and the extent of conservation, propensities, and their contribution to the stability of protein-protein interaction varies substantially between these residues. The case study analyses showed that interface residues, unique to distance, participate in crucial interactions that hold the proteins together, whereas the interface residues unique to the ASA method have a potential role in the recognition, dynamics, and specificity of the complex and can also be a hotspot. Overall, the study recommends applying both distance and ASA methods so that some interface residues missed by either method but crucial to the stability, recognition, dynamics, and function of protein-protein complexes are identified in a complementary manner.

Inhibitors of bacterial immune systems: discovery, mechanisms and applications

To contend with the diversity and ubiquity of bacteriophages and other mobile genetic elements, bacteria have developed an arsenal of immune defence mechanisms. Bacterial defences include CRISPR-Cas, restriction-modification and a growing list of mechanistically diverse systems, which constitute the bacterial 'immune system'. As a response, bacteriophages and mobile genetic elements have evolved direct and indirect mechanisms to circumvent or block bacterial defence pathways and ensure successful infection. Recent advances in methodological and computational approaches, as well as the increasing availability of genome sequences, have boosted the discovery of direct inhibitors of bacterial defence systems. In this Review, we discuss methods for the discovery of direct inhibitors, their diverse mechanisms of action and perspectives on their emerging applications in biotechnology and beyond.

Global Interactome Mapping Reveals Pro-tumorigenic Interactions of NF-κB in Breast Cancer

NF-κB pathway is involved in inflammation; however, recent data shows its role also in cancer development and progression, including metastasis. To understand the role of NF-κB interactome dynamics in cancer, we study the complexity of breast cancer interactome in luminal A breast cancer model and its rearrangement associated with NF-κB modulation. Liquid chromatography-mass spectrometry measurement of 160 size-exclusion chromatography fractions identifies 5460 protein groups. Seven thousand five hundred sixty eight interactions among these proteins have been reconstructed by PrInCE algorithm, of which 2564 have been validated in independent datasets. NF-κB modulation leads to rearrangement of protein complexes involved in NF-κB signaling and immune response, cell cycle regulation, and DNA replication. Central NF-κB transcription regulator RELA co-elutes with interactors of NF-κB activator PRMT5, and these complexes are confirmed by AlphaPulldown prediction. A complementary immunoprecipitation experiment recapitulates RELA interactions with other NF-κB factors, associating NF-κB inhibition with lower binding of NF-κB activators to RELA. This study describes a network of pro-tumorigenic protein interactions and their rearrangement upon NF-κB inhibition with potential therapeutic implications in tumors with high NF-κB activity.

AI-guided pipeline for protein-protein interaction drug discovery identifies a SARS-CoV-2 inhibitor

Protein-protein interactions (PPIs) offer great opportunities to expand the druggable proteome and therapeutically tackle various diseases, but remain challenging targets for drug discovery. Here, we provide a comprehensive pipeline that combines experimental and computational tools to identify and validate PPI targets and perform early-stage drug discovery. We have developed a machine learning approach that prioritizes interactions by analyzing quantitative data from binary PPI assays or AlphaFold-Multimer predictions. Using the quantitative assay LuTHy together with our machine learning algorithm, we identified high-confidence interactions among SARS-CoV-2 proteins for which we predicted three-dimensional structures using AlphaFold-Multimer. We employed VirtualFlow to target the contact interface of the NSP10-NSP16 SARS-CoV-2 methyltransferase complex by ultra-large virtual drug screening. Thereby, we identified a compound that binds to NSP10 and inhibits its interaction with NSP16, while also disrupting the methyltransferase activity of the complex, and SARS-CoV-2 replication. Overall, this pipeline will help to prioritize PPI targets to accelerate the discovery of early-stage drug candidates targeting protein complexes and pathways.

Deep learning in modeling protein complex structures: From contact prediction to end-to-end approaches

Protein-protein interactions play crucial roles in many biological processes. Traditionally, protein complex structures are normally built by protein-protein docking. With the rapid development of artificial intelligence and its great success in monomer protein structure prediction, deep learning has widely been applied to modeling protein-protein complex structures through inter-protein contact prediction and end-to-end approaches in the past few years. This article reviews the recent advances of deep-learning-based approaches in modeling protein-protein complex structures as well as their advantages and limitations. Challenges and possible future directions are also briefly discussed in applying deep learning for the prediction of protein complex structures.

Revolutionizing protein-protein interaction prediction with deep learning

Protein-protein interactions (PPIs) are pivotal for driving diverse biological processes, and any disturbance in these interactions can lead to disease. Thus, the study of PPIs has been a central focus in biology. Recent developments in deep learning methods, coupled with the vast genomic sequence data, have significantly boosted the accuracy of predicting protein structures and modeling protein complexes, approaching levels comparable to experimental techniques. Herein, we review the latest advances in the computational methods for modeling 3D protein complexes and the prediction of protein interaction partners, emphasizing the application of deep learning methods deriving from coevolution analysis. The review also highlights biomedical applications of PPI prediction and outlines challenges in the field.

3Dmapper: a command line tool for BioBank-scale mapping of variants to protein structures

MOTIVATION

The interpretation of genomic data is crucial to understand the molecular mechanisms of biological processes. Protein structures play a vital role in facilitating this interpretation by providing functional context to genetic coding variants. However, mapping genes to proteins is a tedious and error-prone task due to inconsistencies in data formats. Over the past two decades, numerous tools and databases have been developed to automatically map annotated positions and variants to protein structures. However, most of these tools are web-based and not well-suited for large-scale genomic data analysis.

RESULTS

To address this issue, we introduce 3Dmapper, a stand-alone command-line tool developed in Python and R. It systematically maps annotated protein positions and variants to protein structures, providing a solution that is both efficient and reliable.

AVAILABILITY AND IMPLEMENTATION

https://github.com/vicruiser/3Dmapper.

Graph-pMHC: graph neural network approach to MHC class II peptide presentation and antibody immunogenicity

Antigen presentation on MHC class II (pMHCII presentation) plays an essential role in the adaptive immune response to extracellular pathogens and cancerous cells. But it can also reduce the efficacy of large-molecule drugs by triggering an anti-drug response. Significant progress has been made in pMHCII presentation modeling due to the collection of large-scale pMHC mass spectrometry datasets (ligandomes) and advances in machine learning. Here, we develop graph-pMHC, a graph neural network approach to predict pMHCII presentation. We derive adjacency matrices for pMHCII using Alphafold2-multimer and address the peptide-MHC binding groove alignment problem with a simple graph enumeration strategy. We demonstrate that graph-pMHC dramatically outperforms methods with suboptimal inductive biases, such as the multilayer-perceptron-based NetMHCIIpan-4.0 (+20.17% absolute average precision). Finally, we create an antibody drug immunogenicity dataset from clinical trial data and develop a method for measuring anti-antibody immunogenicity risk using pMHCII presentation models. Our model increases receiver operating characteristic curve (ROC)-area under the ROC curve (AUC) by 2.57% compared to just filtering peptides by hits in OASis alone for predicting antibody drug immunogenicity.

HighFold: accurately predicting structures of cyclic peptides and complexes with head-to-tail and disulfide bridge constraints

In recent years, cyclic peptides have emerged as a promising therapeutic modality due to their diverse biological activities. Understanding the structures of these cyclic peptides and their complexes is crucial for unlocking invaluable insights about protein target-cyclic peptide interaction, which can facilitate the development of novel-related drugs. However, conducting experimental observations is time-consuming and expensive. Computer-aided drug design methods are not practical enough in real-world applications. To tackles this challenge, we introduce HighFold, an AlphaFold-derived model in this study. By integrating specific details about the head-to-tail circle and disulfide bridge structures, the HighFold model can accurately predict the structures of cyclic peptides and their complexes. Our model demonstrates superior predictive performance compared to other existing approaches, representing a significant advancement in structure-activity research. The HighFold model is openly accessible at https://github.com/hongliangduan/HighFold.

A comprehensive review of protein-centric predictors for biomolecular interactions: from proteins to nucleic acids and beyond

Proteins interact with diverse ligands to perform a large number of biological functions, such as gene expression and signal transduction. Accurate identification of these protein-ligand interactions is crucial to the understanding of molecular mechanisms and the development of new drugs. However, traditional biological experiments are time-consuming and expensive. With the development of high-throughput technologies, an increasing amount of protein data is available. In the past decades, many computational methods have been developed to predict protein-ligand interactions. Here, we review a comprehensive set of over 160 protein-ligand interaction predictors, which cover protein-protein, protein-nucleic acid, protein-peptide and protein-other ligands (nucleotide, heme, ion) interactions. We have carried out a comprehensive analysis of the above four types of predictors from several significant perspectives, including their inputs, feature profiles, models, availability, etc. The current methods primarily rely on protein sequences, especially utilizing evolutionary information. The significant improvement in predictions is attributed to deep learning methods. Additionally, sequence-based pretrained models and structure-based approaches are emerging as new trends.

Rapid structural analysis of bacterial ribosomes in situ

Rapid structural analysis of purified proteins and their complexes has become increasingly common thanks to key methodological advances in cryo-electron microscopy (cryo-EM) and associated data processing software packages. In contrast, analogous structural analysis in cells via cryo-electron tomography (cryo-ET) remains challenging due to critical technical bottlenecks, including low-throughput sample preparation and imaging, and laborious data processing methods. Here, we describe the development of a rapid in situ cryo-ET sample preparation and data analysis workflow that results in the routine determination of sub-nm resolution ribosomal structures. We apply this workflow to E. coli, producing a 5.8 Å structure of the 70S ribosome from cells in less than 10 days, and we expect this workflow will be widely applicable to related bacterial samples.

HIPPO: HIstogram-based Pseudo-POtential for scoring protein-ssRNA fragment-based docking poses

BACKGROUND

The RNA-Recognition motif (RRM) is a protein domain that binds single-stranded RNA (ssRNA) and is present in as much as 2% of the human genome. Despite this important role in biology, RRM-ssRNA interactions are very challenging to study on the structural level because of the remarkable flexibility of ssRNA. In the absence of atomic-level experimental data, the only method able to predict the 3D structure of protein-ssRNA complexes with any degree of accuracy is ssRNA'TTRACT, an ssRNA fragment-based docking approach using ATTRACT. However, since ATTRACT parameters are not ssRNA-specific and were determined in 2010, there is substantial opportunity for enhancement.

RESULTS

Here we present HIPPO, a composite RRM-ssRNA scoring potential derived analytically from contact frequencies in near-native versus non-native docking models. HIPPO consists of a consensus of four distinct potentials, each extracted from a distinct reference pool of protein-trinucleotide docking decoys. To score a docking pose with one potential, for each pair of RNA-protein coarse-grained bead types, each contact is awarded or penalised according to the relative frequencies of this contact distance range among the correct and incorrect poses of the reference pool. Validated on a fragment-based docking benchmark of 57 experimentally solved RRM-ssRNA complexes, HIPPO achieved a threefold or higher enrichment for half of the fragments, versus only a quarter with the ATTRACT scoring function. In particular, HIPPO drastically improved the chance of very high enrichment (12-fold or higher), a scenario where the incremental modelling of entire ssRNA chains from fragments becomes viable. However, for the latter result, more research is needed to make it directly practically applicable. Regardless, our approach already improves upon the state of the art in RRM-ssRNA modelling and is in principle extendable to other types of protein-nucleic acid interactions.

Engineering Matrix-Free Drug Protein Nanoparticles with Promising Penetration through Biobarriers for Treating Corneal Neovascularization

Protein drugs have been widely used in treating various clinical diseases because of their high specificity, fewer side effects, and favorable therapeutic effect, but they greatly suffer from their weak permeability through tissue barriers, high sensitivity to microenvironments, degradation by proteases, and rapid clearance by the immune system. Herein, we disrupted the standard protocol where protein drugs must be delivered as the cargo via a delivery system and innovatively developed a free entrapping matrix strategy by simply mixing bevacizumab (Beva) with zinc ions to generate Beva-NPs (Beva-Zn2+), where Beva is coordinatively cross-linked by zinc ions with a loading efficiency as high as 99.2% ± 0.41%. This strategy was universal to generating various protein NPs, with different metal ions (Cu2+, Fe3+, Mg2+, Sr2+). The synthetic conditions of Beva-NPs were optimized, and the generated mechanism was investigated in detail. The entrapment, releasing profile, and the bioactivities of released Beva were thoroughly studied. By using in situ doping of the fourth-generation polyamindoamine dendrimer (G4), the Beva-G4-NPs exhibited extended ocular retention and penetration through biobarriers in the anterior segment through transcellular and paracellular pathways, effectively inhibiting corneal neovascularization (CNV) from 91.6 ± 2.03% to 13.5 ± 1.87% in a rat model of CNV. This study contributes to engineering of protein NPs by using a facile strategy for overcoming the weaknesses of protein drugs and protein NPs, such as weak tissue barrier permeability, low encapsulation efficiency, poor loading capacity, and susceptibility to inactivation.