Improved prediction of protein-protein interactions using AlphaFold2

Advances and Challenges in Scoring Functions for RNA-Protein Complex Structure Prediction

RNA-protein complexes play a crucial role in cellular functions, providing insights into cellular mechanisms and potential therapeutic targets. However, experimental determination of these complex structures is often time-consuming and resource-intensive, and it rarely yields high-resolution data. Many computational approaches have been developed to predict RNA-protein complex structures in recent years. Despite these advances, achieving accurate and high-resolution predictions remains a formidable challenge, primarily due to the limitations inherent in current RNA-protein scoring functions. These scoring functions are critical tools for evaluating and interpreting RNA-protein interactions. This review comprehensively explores the latest advancements in scoring functions for RNA-protein docking, delving into the fundamental principles underlying various approaches, including coarse-grained knowledge-based, all-atom knowledge-based, and machine-learning-based methods. We critically evaluate the strengths and limitations of existing scoring functions, providing a detailed performance assessment. Considering the significant progress demonstrated by machine learning techniques, we discuss emerging trends and propose future research directions to enhance the accuracy and efficiency of scoring functions in RNA-protein complex prediction. We aim to inspire the development of more sophisticated and reliable computational tools in this rapidly evolving field.

Anti-symmetric framework for balanced learning of protein-protein interactions

MOTIVATION

Protein-protein interactions (PPIs) are essential for the regulation and facilitation of virtually all biological processes. Computational tools, particularly those based on deep learning, are preferred for the efficient prediction of PPIs. Despite recent progress, two challenges remain unresolved: (i) the imbalanced nature of PPI characteristics is often ignored and (ii) there exists a high computational cost associated with capturing long-range dependencies within protein data, typically exhibiting quadratic complexity relative to the length of the protein sequence.

RESULT

Here, we propose an anti-symmetric graph learning model, BaPPI, for the balanced prediction of PPIs and extrapolation of the involved patterns in PPI network. In BaPPI, the contextualized information of protein data is efficiently handled by an attention-free mechanism formed by recurrent convolution operator. The anti-symmetric graph convolutional network is employed to model the uneven distribution within PPI networks, aiming to learn a more robust and balanced representation of the relationships between proteins. Ultimately, the model is updated using asymmetric loss. The experimental results on classical baseline datasets demonstrate that BaPPI outperforms four state-of-the-art PPI prediction methods. In terms of Micro-F1, BaPPI exceeds the second-best method by 6.5% on SHS27K and 5.3% on SHS148K. Further analysis of the generalization ability and patterns of predicted PPIs also demonstrates our model's generalizability and robustness to the imbalanced nature of PPI datasets.

AVAILABILITY AND IMPLEMENTATION

The source code of this work is publicly available at https://github.com/ttan6729/BaPPI.

Large language models in plant biology

Large language models (LLMs), such as ChatGPT, have taken the world by storm. However, LLMs are not limited to human language and can be used to analyze sequential data, such as DNA, protein, and gene expression. The resulting foundation models can be repurposed to identify the complex patterns within the data, resulting in powerful, multipurpose prediction tools able to predict the state of cellular systems. This review outlines the different types of LLMs and showcases their recent uses in biology. Since LLMs have not yet been embraced by the plant community, we also cover how these models can be deployed for the plant kingdom.

Prediction of protein structure and AI

AlphaFold, an artificial intelligence (AI)-based tool for predicting the 3D structure of proteins, is now widely recognized for its high accuracy and versatility in the folding of human proteins. AlphaFold is useful for understanding structure-function relationships from protein 3D structure models and can serve as a template or a reference for experimental structural analysis including X-ray crystallography, NMR and cryo-EM analysis. Its use is expanding among researchers, not only in structural biology but also in other research fields. Researchers are currently exploring the full potential of AlphaFold-generated protein models. Predicting disease severity caused by missense mutations is one such application. This article provides an overview of the 3D structural modeling of AlphaFold based on deep learning techniques and highlights the challenges in predicting the pathogenicity of missense mutations.

Exploring the shared pathogenic strategies of independently evolved effectors across distinct plant viruses

Plants have developed very diverse strategies to defend themselves against viral pathogens, among which plant hormones play pivotal roles. In response, some viruses have also deployed multifunctional viral effectors that effectively hijack key component hubs to counter or evade plant immune surveillance. Although significant progress has been made toward understanding counter-defense strategies that manipulate plant hormone regulatory molecules, these efforts have often been limited to an individual virus or specific host target/pathway. This review provides new insights into broad-spectrum antiviral responses in rice triggered by key components of phytohormone signaling, and highlights the common features of counter-defense strategies employed by distinct rice-infecting RNA viruses. These strategies involve the secretion of multifunctional virulence effectors that target the sophisticated phytohormone system, dampening immune responses by engaging with the same host targets. Additionally, the review provides an in-depth exploration of various viral effectors, emphasizing tertiary structure-based research and shared host targets. Understanding these conserved characteristics in detail may pave the way for molecular drug design, opening new opportunities to enhance broad-spectrum antiviral trials through precise engineering.

Using AlphaFold Multimer to discover interkingdom protein-protein interactions

Structural prediction by artificial intelligence can be powerful new instruments to discover novel protein-protein interactions, but the community still grapples with the implementation, opportunities and limitations. Here, we discuss and re-analyse our in silico screen for novel pathogen-secreted inhibitors of immune hydrolases to illustrate the power and limitations of structural predictions. We discuss strategies of curating sequences, including controls, and reusing sequence alignments and highlight important limitations caused by different platforms, sequence depth and computing times. We hope these experiences will support similar interactomic screens by the research community.

Insights into AlphaFold's breakthrough in neurodegenerative diseases

Neurodegenerative diseases (ND) are disorders of the central nervous system (CNS) characterized by impairment in neurons' functions, and complete loss, leading to memory loss, and difficulty in learning, language, and movement processes. The most common among these NDs are Alzheimer's disease (AD) and Parkinson's disease (PD), although several other disorders also exist. These are frontotemporal dementia (FTD), amyotrophic lateral syndrome (ALS), Huntington's disease (HD), and others; the major pathological hallmark of NDs is the proteinopathies, either of amyloid-β (Aβ), tauopathies, or synucleinopathies. Aggregation of proteins that do not undergo normal configuration, either due to mutations or through some disturbance in cellular pathway contributes to the diseases. Artificial Intelligence (AI) and deep learning (DL) have proven to be successful in the diagnosis and treatment of various congenital diseases. DL approaches like AlphaFold (AF) are a major leap towards success in CNS disorders. This 3D protein geometry modeling algorithm developed by DeepMind has the potential to revolutionize biology. AF has the potential to predict 3D-protein confirmation at an accuracy level comparable to experimentally predicted one, with the additional advantage of precisely estimating protein interactions. This breakthrough will be beneficial to identify diseases' advancement and the disturbance of signaling pathways stimulating impaired functions of proteins. Though AlphaFold has solved a major problem in structural biology, it cannot predict membrane proteins-a beneficial approach for drug designing.

Computational tools to predict context-specific protein complexes

Interactions between thousands of proteins define cells' protein-protein interaction (PPI) network. Some of these interactions lead to the formation of protein complexes. It is challenging to identify a protein complex in a haystack of protein-protein interactions, and it is even more difficult to predict all protein complexes of the complexome. Simulations and machine learning approaches try to crack these problems by looking at the PPI network or predicted protein structures. Clustering of PPI networks led to the first protein complex predictions, while most recently, atomistic models of protein complexes and deep-learning-based structure prediction methods have also emerged. The simulation of PPI level interactions even enables the quantitative prediction of protein complexes. These methods, the required data sources, and their potential future developments are discussed in this review.

A virally encoded tRNA neutralizes the PARIS antiviral defence system

Viruses compete with each other for limited cellular resources, and some deliver defence mechanisms that protect the host from competing genetic parasites1. The phage antirestriction induced system (PARIS) is a defence system, often encoded in viral genomes, that is composed of a 55 kDa ABC ATPase (AriA) and a 35 kDa TOPRIM nuclease (AriB)2. However, the mechanism by which AriA and AriB function in phage defence is unknown. Here we show that AriA and AriB assemble into a 425 kDa supramolecular immune complex. We use cryo-electron microscopy to determine the structure of this complex, thereby explaining how six molecules of AriA assemble into a propeller-shaped scaffold that coordinates three subunits of AriB. ATP-dependent detection of foreign proteins triggers the release of AriB, which assembles into a homodimeric nuclease that blocks infection by cleaving host lysine transfer RNA. Phage T5 subverts PARIS immunity through expression of a lysine transfer RNA variant that is not cleaved by PARIS, thereby restoring viral infection. Collectively, these data explain how AriA functions as an ATP-dependent sensor that detects viral proteins and activates the AriB toxin. PARIS is one of an emerging set of immune systems that form macromolecular complexes for the recognition of foreign proteins, rather than foreign nucleic acids3.

Autoinhibition of ubiquitin-specific protease 8: Insights into domain interactions and mechanisms of regulation

Ubiquitin-specific proteases (USPs) are a family of multi-domain deubiquitinases (DUBs) with variable architectures, some containing regulatory auxiliary domains. Among the USP family, all occurrences of intramolecular regulation presently known are autoactivating. USP8 remains the sole exception as its putative WW-like domain, conserved only in vertebrate orthologs, is autoinhibitory. Here, we present a comprehensive structure-function analysis describing the autoinhibition of USP8 and provide evidence of the physical interaction between the WW-like and catalytic domains. The solution structure of full-length USP8 reveals an extended, monomeric conformation. Coupled with DUB assays, the WW-like domain is confirmed to be the minimal autoinhibitory unit. Strikingly, autoinhibition is only observed with the WW-like domain in cis and depends on the length of the linker tethering it to the catalytic domain. Modeling of the WW:CD complex structure and mutagenesis of interface residues suggests a novel binding site in the S1 pocket. To investigate the interplay between phosphorylation and USP8 autoinhibition, we identify AMP-activated protein kinase as a highly selective modifier of S718 in the 14-3-3 binding motif. We show that 14-3-3γ binding to phosphorylated USP8 potentiates autoinhibition in a WW-like domain-dependent manner by stabilizing an autoinhibited conformation. These findings provide mechanistic details on the autoregulation of USP8 and shed light on its evolutionary significance.

Allergies to Allergens from Cats and Dogs: A Review and Update on Sources, Pathogenesis, and Strategies

Inhalation allergies caused by cats and dogs can lead to a range of discomforting symptoms, such as rhinitis and asthma, in humans. With the increasing popularity of and care provided to these companion animals, the allergens they produce pose a growing threat to susceptible patients' health. Allergens from cats and dogs have emerged as significant risk factors for triggering asthma and allergic rhinitis worldwide; however, there remains a lack of systematic measures aimed at assisting individuals in recognizing and preventing allergies caused by these animals. This review provides comprehensive insights into the classification of cat and dog allergens, along with their pathogenic mechanisms. This study also discusses implementation strategies for prevention and control measures, including physical methods, gene-editing technology, and immunological approaches, as well as potential strategies for enhancing allergen immunotherapy combined with immunoinformatics. Finally, it presents future prospects for the prevention and treatment of human allergies caused by cats and dogs. This review will improve knowledge regarding allergies to cats and dogs while providing insights into potential targets for the development of next-generation treatments.

Revolutionizing Molecular Design for Innovative Therapeutic Applications through Artificial Intelligence

The field of computational protein engineering has been transformed by recent advancements in machine learning, artificial intelligence, and molecular modeling, enabling the design of proteins with unprecedented precision and functionality. Computational methods now play a crucial role in enhancing the stability, activity, and specificity of proteins for diverse applications in biotechnology and medicine. Techniques such as deep learning, reinforcement learning, and transfer learning have dramatically improved protein structure prediction, optimization of binding affinities, and enzyme design. These innovations have streamlined the process of protein engineering by allowing the rapid generation of targeted libraries, reducing experimental sampling, and enabling the rational design of proteins with tailored properties. Furthermore, the integration of computational approaches with high-throughput experimental techniques has facilitated the development of multifunctional proteins and novel therapeutics. However, challenges remain in bridging the gap between computational predictions and experimental validation and in addressing ethical concerns related to AI-driven protein design. This review provides a comprehensive overview of the current state and future directions of computational methods in protein engineering, emphasizing their transformative potential in creating next-generation biologics and advancing synthetic biology.

Systematic discovery of DNA-binding tandem repeat proteins

Tandem repeat proteins (TRPs) are widely distributed and bind to a wide variety of ligands. DNA-binding TRPs such as zinc finger (ZNF) and transcription activator-like effector (TALE) play important roles in biology and biotechnology. In this study, we first conducted an extensive analysis of TRPs in public databases, and found that the enormous diversity of TRPs is largely unexplored. We then focused our efforts on identifying novel TRPs possessing DNA-binding capabilities. We established a protein language model for DNA-binding protein prediction (PLM-DBPPred), and predicted a large number of DNA-binding TRPs. A subset was then selected for experimental screening, leading to the identification of 11 novel DNA-binding TRPs, with six showing sequence specificity. Notably, members of the STAR (Short TALE-like Repeat proteins) family can be programmed to target specific 9 bp DNA sequences with high affinity. Leveraging this property, we generated artificial transcription factors using reprogrammed STAR proteins and achieved targeted activation of endogenous gene sets. Furthermore, the members of novel families such as MOON (Marine Organism-Originated DNA binding protein) and pTERF (prokaryotic mTERF-like protein) exhibit unique features and distinct DNA-binding characteristics, revealing interesting biological clues. Our study expands the diversity of DNA-binding TRPs, and demonstrates that a systematic approach greatly enhances the discovery of new biological insights and tools.

RudS: bacterial desulfidase responsible for tRNA 4-thiouridine de-modification

In this study, we present an extensive analysis of a widespread group of bacterial tRNA de-modifying enzymes, dubbed RudS, which consist of a TudS desulfidase fused to a Domain of Unknown Function 1722 (DUF1722). RudS enzymes exhibit specific de-modification activity towards the 4-thiouridine modification (s4U) in tRNA molecules, as indicated by our experimental findings. The heterologous overexpression of RudS genes in Escherichia coli significantly reduces the tRNA 4-thiouridine content and diminishes UVA-induced growth delay, indicating the enzyme's role in regulating photosensitive tRNA s4U modification. Through a combination of protein modeling, docking studies, and molecular dynamics simulations, we have identified amino acid residues involved in catalysis and tRNA binding. Experimental validation through targeted mutagenesis confirms the TudS domain as the catalytic core of RudS, with the DUF1722 domain facilitating tRNA binding in the anticodon region. Our results suggest that RudS tRNA modification eraser proteins may play a role in regulating tRNA during prokaryotic stress responses.

Integrative Modeling of Protein-Polypeptide Complexes by Bayesian Model Selection using AlphaFold and NMR Chemical Shift Perturbation Data

Protein-polypeptide interactions, including those involving intrinsically-disordered peptides and intrinsically-disordered regions of protein binding partners, are crucial for many biological functions. However, experimental structure determination of protein-peptide complexes can be challenging. Computational methods, while promising, generally require experimental data for validation and refinement. Here we present CSP_Rank, an integrated modeling approach to determine the structures of protein-peptide complexes. This method combines AlphaFold2 (AF2) enhanced sampling methods with a Bayesian conformational selection process based on experimental Nuclear Magnetic Resonance (NMR) Chemical Shift Perturbation (CSP) data and AF2 confidence metrics. Using a curated dataset of 108 protein-peptide complexes from the Biological Magnetic Resonance Data Bank (BMRB), we observe that while AF2 typically yields models with excellent consistency with experimental CSP data, applying enhanced sampling followed by data-guided conformational selection routinely results in ensembles of structures with improved agreement with NMR observables. For two systems, we cross-validate the CSP-selected models using independently acquired nuclear Overhauser effect (NOE) NMR data and demonstrate how CSP and NMR can be combined using our Bayesian framework for model selection. CSP_Rank is a novel method for integrative modeling of protein-peptide complexes and has broad implications for studies of protein-peptide interactions and aiding in understanding their biological functions.

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

Type-A γ-aminobutyric acid (GABAA) receptors are channel proteins crucial to mediating neuronal balance in the central nervous system (CNS). The structure of GABAA receptors allows for multiple binding sites and is key to drug development. Yet the formation mechanism of the receptor's distinctive pentameric structure is still unknown. This study aims to investigate the role of three predominant subunits of the human GABAA receptor in the formation of protein pentamers. Through purifying and refolding the protein fragments of the GABAA receptor α1, β2, and γ2 subunits, the particle structures were visualised with negative staining electron microscopy (EM). To aid the analysis, AlphaFold2 was used to compare the structures. Results show that α1 and β2 subunit fragments successfully formed homo-oligomers, particularly homopentameric structures, while the predominant heteropentameric GABAA receptor was also replicated through the combination of the three subunits. However, homopentameric structures were not observed with the γ2 subunit proteins. A comparison of the AlphaFold2 predictions and the previously obtained cryo-EM structures presents new insights into the subunits' modular structure and polymerization status. By performing experimental and computational studies, a deeper understanding of the complex structure of GABAA receptors is provided. Hopefully, this study can pave the way to developing novel therapeutics for neuropsychiatric diseases.

Computational modeling of mast cell tryptase family informs selective inhibitor development

Mast cell tryptases, a family of serine proteases involved in inflammatory responses and cancer development, present challenges in structural characterization and inhibitor development. We employed state-of-the-art protein structure prediction algorithms to model the three-dimensional structures of tryptases α, β, δ, γ, and ε with high accuracy. Computational docking identified potential substrates and inhibitors, suggesting overlapping yet distinct activities. Tryptases β, δ, and ε were predicted to act on phenolic compounds, with β and ε additionally hydrolyzing cyanides. Tryptase δ may possess unique formyl-CoA dehydrogenase activity. Virtual screening revealed 63 compounds exhibiting strong binding to tryptase β (TPSB2), 12 exceeding the affinity of the known inhibitor. Notably, the top hit (3-chloro-4-methylbenzimidamide) displayed over 10-fold selectivity for tryptase β over other isoforms. Our integrative approach combining protein modeling, functional annotation, and molecular docking provides a framework for characterizing tryptase isoforms and developing selective inhibitors of therapeutic potential in inflammatory and cancer conditions.

Rational Design and Modification of NphB for Cannabinoids Biosynthesis

The rapidly growing field of cannabinoid research is gaining recognition for its impact in neuropsychopharmacology and mood regulation. However, prenyltransferase (NphB) (a key enzyme in cannabinoid precursor synthesis) still needs significant improvement in order to be usable in large-scale industrial applications due to low activity and limited product range. By rational design and high-throughput screening, NphB's catalytic efficiency and product diversity have been markedly enhanced, enabling direct production of a range of cannabinoids, without the need for traditional enzymatic conversions, thus broadening the production scope of cannabinoids, including cannabigerol (CBG), cannabigerolic acid (CBGA), cannabigerovarin (CBGV), and cannabigerovarinic acid (CBGVA). Notably, the W3 mutant achieved a 10.6-fold increase in CBG yield and exhibited a 10.3- and 20.8-fold enhancement in catalytic efficiency for CBGA and CBGV production, respectively. The W4 mutant also displayed an 9.3-fold increase in CBGVA activity. Molecular dynamics simulations revealed that strategic reconfiguration of the active site's hydrogen bonding network, disulfide bond formation, and enhanced hydrophobic interactions are pivotal for the improved synthetic efficiency of these NphB mutants. Our findings advance the understanding of enzyme optimization for cannabinoid synthesis and lay a foundation for the industrial-scale production of these valuable compounds.

High-throughput discovery of inhibitory protein fragments with AlphaFold

Peptides can bind to specific sites on larger proteins and thereby function as inhibitors and regulatory elements. Peptide fragments of larger proteins are particularly attractive for achieving these functions due to their inherent potential to form native-like binding interactions. Recently developed experimental approaches allow for high-throughput measurement of protein fragment inhibitory activity in living cells. However, it has thus far not been possible to predict de novo which of the many possible protein fragments bind to protein targets, let alone act as inhibitors. We have developed a computational method, FragFold, that employs AlphaFold to predict protein fragment binding to full-length proteins in a high-throughput manner. Applying FragFold to thousands of fragments tiling across diverse proteins revealed peaks of predicted binding along each protein sequence. Comparisons with experimental measurements establish that our approach is a sensitive predictor of fragment function: Evaluating inhibitory fragments from known protein-protein interaction interfaces, we find 87% are predicted by FragFold to bind in a native-like mode. Across full protein sequences, 68% of FragFold-predicted binding peaks match experimentally measured inhibitory peaks. Deep mutational scanning experiments support the predicted binding modes and uncover superior inhibitory peptides in high throughput. Further, FragFold is able to predict previously unknown protein binding modes, explaining prior genetic and biochemical data. The success rate of FragFold demonstrates that this computational approach should be broadly applicable for discovering inhibitory protein fragments across proteomes.

SymProFold: Structural prediction of symmetrical biological assemblies

Symmetry in nature often emerges from self-assembly processes and serves a wide range of functions. Cell surface layers (S-layers) form symmetrical lattices on many bacterial and archaeal cells, playing essential roles such as facilitating cell adhesion, evading the immune system, and protecting against environmental stress. However, the experimental structural characterization of these S-layers is challenging due to their self-assembly properties and high sequence variability. In this study, we introduce the SymProFold pipeline, which utilizes the high accuracy of AlphaFold-Multimer predictions to derive symmetrical assemblies from protein sequences, specifically focusing on two-dimensional S-layer arrays and spherical viral capsids. The pipeline tests all known symmetry operations observed in these systems (p1, p2, p3, p4, and p6) and identifies the most likely symmetry for the assembly. The predicted models were validated using available experimental data at the cellular level, and additional crystal structures were obtained to confirm the symmetry and interfaces of several SymProFold assemblies. Overall, the SymProFold pipeline enables the determination of symmetric protein assemblies linked to critical functions, thereby opening possibilities for exploring functionalities and designing targeted applications in diverse fields such as nanotechnology, biotechnology, medicine, and materials and environmental sciences.

Unlocking protein-protein interactions in plants: a comprehensive review of established and emerging techniques

Protein-protein interactions orchestrate plant development and serve as crucial elements for cellular and environmental communication. Understanding these interactions offers a gateway to unravel complex protein networks that will allow a better understanding of nature. Methods for the characterization of protein-protein interactions have been around over 30 years, yet the complexity of some of these interactions has fueled the development of new techniques that provide a better understanding of the underlying dynamics. In many cases, the application of these techniques is limited by the nature of the available sample. While some methods require an in vivo set-up, others solely depend on protein sequences to study protein-protein interactions via an in silico set-up. The vast number of techniques available to date calls for a way to select the appropriate tools for the study of specific interactions. Here, we classify widely spread tools and new emerging techniques for the characterization of protein-protein interactions based on sample requirements while providing insights into the information that they can potentially deliver. We provide a comprehensive overview of commonly used techniques and elaborate on the most recent developments, showcasing their implementation in plant research.

Crucial amino acids identified in Δ12 fatty acid desaturases related to linoleic acid production in Perilla frutescens

Perilla oil from the medicinal crop Perilla frutescens possess a wide range of biological activities and is generally used as an edible oil in many countries. The molecular basis for its formation is of particular relevance to perilla and its breeders. Here in the present study, four PfFAD2 genes were identified in different perilla cultivars, PF40 and PF70, with distinct oil content levels, respectively. Their function was characterized in engineered yeast strain, and among them, PfFAD2-1PF40, PfFAD2-1PF70 had no LA biosynthesis ability, while PfFAD2-2PF40 in cultivar with high oil content levels possessed higher catalytic activity than PfFAD2-2PF70. Key amino acid residues responsible for the enhanced catalytic activity of PfFAD2-2PF40 was identified as residue R221 through sequence alignment, molecular docking, and site-directed mutation studies. Moreover, another four amino acid residues influencing PfFAD2 catalytic activity were discovered through random mutation analysis. This study lays a theoretical foundation for the genetic improvement of high-oil-content perilla cultivars and the biosynthesis of LA and its derivatives.

Protist biochemistry: Functional SUF system facilitated mitochondrial loss

The replacement of the mitochondrial iron-sulfur cluster machinery by a cytosolic system may have triggered complete mitochondrial loss in the anaerobic protist Monocercomonoides exilis. A new study strongly supports this scenario by confirming the functionality of this system.

Artificial intelligence for geoscience: Progress, challenges, and perspectives

This paper explores the evolution of geoscientific inquiry, tracing the progression from traditional physics-based models to modern data-driven approaches facilitated by significant advancements in artificial intelligence (AI) and data collection techniques. Traditional models, which are grounded in physical and numerical frameworks, provide robust explanations by explicitly reconstructing underlying physical processes. However, their limitations in comprehensively capturing Earth's complexities and uncertainties pose challenges in optimization and real-world applicability. In contrast, contemporary data-driven models, particularly those utilizing machine learning (ML) and deep learning (DL), leverage extensive geoscience data to glean insights without requiring exhaustive theoretical knowledge. ML techniques have shown promise in addressing Earth science-related questions. Nevertheless, challenges such as data scarcity, computational demands, data privacy concerns, and the "black-box" nature of AI models hinder their seamless integration into geoscience. The integration of physics-based and data-driven methodologies into hybrid models presents an alternative paradigm. These models, which incorporate domain knowledge to guide AI methodologies, demonstrate enhanced efficiency and performance with reduced training data requirements. This review provides a comprehensive overview of geoscientific research paradigms, emphasizing untapped opportunities at the intersection of advanced AI techniques and geoscience. It examines major methodologies, showcases advances in large-scale models, and discusses the challenges and prospects that will shape the future landscape of AI in geoscience. The paper outlines a dynamic field ripe with possibilities, poised to unlock new understandings of Earth's complexities and further advance geoscience exploration.

AlphaCRV: a pipeline for identifying accurate binder topologies in mass-modeling with AlphaFold

Motivation

The speed and accuracy of deep learning-based structure prediction algorithms make it now possible to perform in silico "pull-downs" to identify protein-protein interactions on a proteome-wide scale. However, on such a large scale, existing scoring algorithms are often insufficient to discriminate biologically relevant interactions from false positives.

Results

Here, we introduce AlphaCRV, a Python package that helps identify correct interactors in a one-against-many AlphaFold screen by clustering, ranking, and visualizing conserved binding topologies, based on protein sequence and fold.

Availability and implementation

AlphaCRV is a Python package for Linux, freely available at https://github.com/strubelab/AlphaCRV.

USP37 prevents premature disassembly of stressed replisomes by TRAIP

The E3 ubiquitin ligase TRAIP associates with the replisome and helps this molecular machine deal with replication stress. Thus, TRAIP promotes DNA inter-strand crosslink repair by triggering the disassembly of CDC45-MCM2-7-GINS (CMG) helicases that have converged on these lesions. However, disassembly of single CMGs that have stalled temporarily would be deleterious, suggesting that TRAIP must be carefully regulated. Here, we demonstrate that human cells lacking the de-ubiquitylating enzyme USP37 are hypersensitive to topoisomerase poisons and other replication stress-inducing agents. We further show that TRAIP loss rescues the hypersensitivity of USP37 knockout cells to topoisomerase inhibitors. In Xenopus egg extracts depleted of USP37, TRAIP promotes premature CMG ubiquitylation and disassembly when converging replisomes stall. Finally, guided by AlphaFold-Multimer, we discovered that binding to CDC45 mediates USP37's response to topological stress. In conclusion, we propose that USP37 protects genome stability by preventing TRAIP-dependent CMG unloading when replication stress impedes timely termination.

SEPALLATA-driven MADS transcription factor tetramerization is required for inner whorl floral organ development

MADS transcription factors are master regulators of plant reproduction and flower development. The SEPALLATA (SEP) subfamily of MADS transcription factors is required for the development of floral organs and plays roles in inflorescence architecture and development of the floral meristem. SEPALLATAs act as organizers of MADS complexes, forming both heterodimers and heterotetramers in vitro. To date, the MADS complexes characterized in angiosperm floral organ development contain at least 1 SEPALLATA protein. Whether DNA binding by SEPALLATA-containing dimeric MADS complexes is sufficient for launching floral organ identity programs, however, is not clear as only defects in floral meristem determinacy were observed in tetramerization-impaired SEPALLATA mutant proteins. Here, we used a combination of genome-wide-binding studies, high-resolution structural studies of the SEP3/AGAMOUS (AG) tetramerization domain, structure-based mutagenesis and complementation experiments in Arabidopsis (Arabidopsis thaliana) sep1 sep2 sep3 and sep1 sep2 sep3 ag-4 plants transformed with versions of SEP3 encoding tetramerization mutants. We demonstrate that while SEP3 heterodimers can bind DNA both in vitro and in vivo and recognize the majority of SEP3 wild-type-binding sites genome-wide, tetramerization is required not only for floral meristem determinacy but also for floral organ identity in the second, third, and fourth whorls.

Bound ion effects: Using machine learning method to study the kinesin Ncd's binding with microtubule

Drosophila Ncd proteins are motor proteins that play important roles in spindle organization. Ncd and the tubulin dimer are highly charged. Thus, it is crucial to investigate Ncd-tubulin dimer interactions in the presence of ions, especially ions that are bound or restricted at the Ncd-tubulin dimer binding interfaces. To consider the ion effects, widely used implicit solvent models treat ions implicitly in the continuous solvent environment without focusing on the individual ions' effects. But highly charged biomolecules such as the Ncd and tubulin dimer may capture some ions at highly charged regions as bound ions. Such bound ions are restricted to their binding sites; thus, they can be treated as part of the biomolecules. By applying multiscale computational methods, including the machine-learning-based Hybridizing Ions Treatment-2 program, molecular dynamics simulations, DelPhi, and DelPhiForce, we studied the interaction between the Ncd motor domain and the tubulin dimer using a hybrid solvent model, which considers the bound ions explicitly and the other ions implicitly in the solvent environment. To identify the importance of treating bound ions explicitly, we also performed calculations using the implicit solvent model without considering the individual bound ions. We found that the calculations of the electrostatic features differ significantly between those of the hybrid solvent model and the pure implicit solvent model. The analyses show that treating bound ions at highly charged regions explicitly is crucial for electrostatic calculations. This work proposes a machine-learning-based approach to handle the bound ions using the hybrid solvent model. Such an approach is not only capable of handling kinesin-tubulin complexes but is also appropriate for other highly charged biomolecules, such as DNA/RNA, viral capsid proteins, etc.

Polypharmacology prediction: the long road toward comprehensively anticipating small-molecule selectivity to de-risk drug discovery

INTRODUCTION

Small molecules often bind to multiple targets, a behavior termed polypharmacology. Anticipating polypharmacology is essential for drug discovery since unknown off-targets can modulate safety and efficacy - profoundly affecting drug discovery success. Unfortunately, experimental methods to assess selectivity present significant limitations and drugs still fail in the clinic due to unanticipated off-targets. Computational methods are a cost-effective, complementary approach to predict polypharmacology.

AREAS COVERED

This review aims to provide a comprehensive overview of the state of polypharmacology prediction and discuss its strengths and limitations, covering both classical cheminformatics methods and bioinformatic approaches. The authors review available data sources, paying close attention to their different coverage. The authors then discuss major algorithms grouped by the types of data that they exploit using selected examples.

EXPERT OPINION

Polypharmacology prediction has made impressive progress over the last decades and contributed to identify many off-targets. However, data incompleteness currently limits most approaches to comprehensively predict selectivity. Moreover, our limited agreement on model assessment challenges the identification of the best algorithms - which at present show modest performance in prospective real-world applications. Despite these limitations, the exponential increase of multidisciplinary Big Data and AI hold much potential to better polypharmacology prediction and de-risk drug discovery.

Nonstructural protein 4 of human norovirus self-assembles into various membrane-bridging multimers

Single-stranded, positive-sense RNA ((+)RNA) viruses replicate their genomes in virus-induced intracellular membrane compartments. (+)RNA viruses dedicate a significant part of their small genomes (a few thousands to a few tens of thousands of bases) to the generation of these compartments by encoding membrane-interacting proteins and/or protein domains. Noroviruses are a very diverse genus of (+)RNA viruses including human and animal pathogens. Human noroviruses are the major cause of acute gastroenteritis worldwide, with genogroup II genotype 4 (GII.4) noroviruses accounting for the vast majority of infections. Three viral proteins encoded in the N terminus of the viral replication polyprotein direct intracellular membrane rearrangements associated with norovirus replication. Of these three, nonstructural protein 4 (NS4) seems to be the most important, although its exact functions in replication organelle formation are unknown. Here, we produce, purify, and characterize GII.4 NS4. AlphaFold modeling combined with experimental data refines and corrects our previous crude structural model of NS4. Using simple artificial liposomes, we report an extensive characterization of the membrane properties of NS4. We find that NS4 self-assembles and thereby bridges liposomes together. Cryo-EM, NMR, and membrane flotation show formation of several distinct NS4 assemblies, at least two of them bridging pairs of membranes together in different fashions. Noroviruses belong to (+)RNA viruses whose replication compartment is extruded from the target endomembrane and generates double-membrane vesicles. Our data establish that the 21-kDa GII.4 human norovirus NS4 can, in the absence of any other factor, recapitulate in tubo several features, including membrane apposition, that occur in such processes.

Discovery of potential antidiabetic peptides using deep learning

Antidiabetic peptides (ADPs), peptides with potential antidiabetic activity, hold significant importance in the treatment and control of diabetes. Despite their therapeutic potential, the discovery and prediction of ADPs remain challenging due to limited data, the complex nature of peptide functions, and the expensive and time-consuming nature of traditional wet lab experiments. This study aims to address these challenges by exploring methods for the discovery and prediction of ADPs using advanced deep learning techniques. Specifically, we developed two models: a single-channel CNN and a three-channel neural network (CNN + RNN + Bi-LSTM). ADPs were primarily gathered from the BioDADPep database, alongside thousands of non-ADPs sourced from anticancer, antibacterial, and antiviral peptide datasets. Subsequently, data preprocessing was performed with the evolutionary scale model (ESM-2), followed by model training and evaluation through 10-fold cross-validation. Furthermore, this work collected a series of newly published ADPs as an independent test set through literature review, and found that the CNN model achieved the highest accuracy (90.48 %) in predicting the independent test set, surpassing existing ADP prediction tools. Finally, the application of the model was considered. SeqGAN was used to generate new candidate ADPs, followed by screening with the constructed CNN model. Selected peptides were then evaluated using physicochemical property prediction and structural forecasts for pharmaceutical potential. In summary, this study not only established robust ADP prediction models but also employed these models to screen a batch of potential ADPs, addressing a critical need in the field of peptide-based antidiabetic research.

Monoclonal antibody development for early detection of ASFV I73R protein: Identification of a linear antigenic epitope

African swine fever virus (ASFV), which was first identified in northern China in 2018, causes high mortality in pigs. Since the I73R protein in ASFV is abundantly expressed during the early phase of virus replication, it can be used as a target protein for early diagnosis. In this study, the I73R protein of ASFV was expressed, and we successfully prepared a novel monoclonal antibody (mAb), 8G11D7, that recognizes this protein. Through both indirect immunofluorescence and Western blotting assays, we demonstrated that 8G11D7 can detect ASFV strains. By evaluating the binding of the antibody to a series of I73R-truncated peptides, the definitive epitope recognized by the monoclonal antibody 8G11D7 was determined to be 58 DKTNTIYPP 66. Bioinformatic analysis revealed that the antigenic epitope had a high antigenic index and conservatism. This study contributes to a deeper understanding of ASFV protein structure and function, helping establish ASFV-specific detection method.

Plastid HSP90C C-terminal extension region plays a regulatory role in chaperone activity and client binding

HSP90Cs are essential molecular chaperones localized in the plastid stroma that maintain protein homeostasis and assist the import and thylakoid transport of chloroplast proteins. While HSP90C contains all conserved domains as an HSP90 family protein, it also possesses a unique feature in its variable C-terminal extension (CTE) region. This study elucidated the specific function of this HSP90C CTE region. Our phylogenetic analyses revealed that this intrinsically disordered region contains a highly conserved DPW motif in the green lineages. With biochemical assays, we showed that the CTE is required for the chaperone to effectively interact with client proteins PsbO1 and LHCB2 to regulate ATP-independent chaperone activity and to effectuate its ATP hydrolysis. The CTE truncation mutants could support plant growth and development reminiscing the wild type under normal conditions except for a minor phenotype in cotyledon when expressed at a level comparable to wild type. However, higher HSP90C expression was observed to correlate with a stronger response to specific photosystem II inhibitor DCMU, and CTE truncations dampened the response. Additionally, when treated with lincomycin to inhibit chloroplast protein translation, CTE truncation mutants showed a delayed response to PsbO1 expression repression, suggesting its role in chloroplast retrograde signaling. Our study therefore provides insights into the mechanism of HSP90C in client protein binding and the regulation of green chloroplast maturation and function, especially under stress conditions.

Structures prediction and replica exchange molecular dynamics simulations of α-synuclein: A case study for intrinsically disordered proteins

In recent years, a variety of three-dimensional structure prediction tools, including AlphaFold2, AlphaFold3, I-TASSER, C-I-TASSER, Phyre2, ESMFold, and RoseTTAFold, have been employed in the investigation of intrinsically disordered proteins. However, a comprehensive validation of these tools specifically for intrinsically disordered proteins has yet to be conducted. In this study, we utilize AlphaFold2, AlphaFold3, I-TASSER, C-I-TASSER, Phyre2, ESMFold, and RoseTTAFold to predict the structure of a model intrinsically disordered α-synuclein protein. Additionally, extensive replica exchange molecular dynamics simulations of the intrinsically disordered protein are conducted. The resulting structures from both structure prediction tools and replica exchange molecular dynamics simulations are analyzed for radius of gyration, secondary and tertiary structure properties, as well as Cα and Hα chemical shift values. A comparison of the obtained results with experimental data reveals that replica exchange molecular dynamics simulations provide results in excellent agreement with experimental observations. However, none of the structure prediction tools utilized in this study can fully capture the structural characteristics of the model intrinsically disordered protein. This study shows that a cluster of ensembles are required for intrinsically disordered proteins. Artificial-intelligence based structure prediction tools such as AlphaFold3 and C-I-TASSER could benefit from stochastic sampling or Monte Carlo simulations for generating an ensemble of structures for intrinsically disordered proteins.

Unveiling the evolution of policies for enhancing protein structure predictions: A comprehensive analysis

Predicting protein structure is both fascinating and formidable, playing a crucial role in structure-based drug discovery and unraveling diseases with elusive origins. The Critical Assessment of Protein Structure Prediction (CASP) serves as a biannual battleground where global scientists converge to untangle the intricate relationships within amino acid chains. Two primary methods, Template-Based Modeling (TBM) and Template-Free (TF) strategies, dominate protein structure prediction. The trend has shifted towards Template-Free predictions due to their broader sequence coverage with fewer templates. The predictive process can be broadly classified into contact map, binned-distance, and real-valued distance predictions, each with distinctive strengths and limitations manifested through tailored loss functions. We have also introduced revolutionary end-to-end, and all-atom diffusion-based techniques that have transformed protein structure predictions. Recent advancements in deep learning techniques have significantly improved prediction accuracy, although the effectiveness is contingent upon the quality of input features derived from natural bio-physiochemical attributes and Multiple Sequence Alignments (MSA). Hence, the generation of high-quality MSA data holds paramount importance in harnessing informative input features for enhanced prediction outcomes. Remarkable successes have been achieved in protein structure prediction accuracy, however not enough for what structural knowledge was intended to, which implies need for development in some other aspects of the predictions. In this regard, scientists have opened other frontiers for protein structural prediction. The utilization of subsampling in multiple sequence alignment (MSA) and protein language modeling appears to be particularly promising in enhancing the accuracy and efficiency of predictions, ultimately aiding in drug discovery efforts. The exploration of predicting protein complex structure also opens up exciting opportunities to deepen our knowledge of molecular interactions and design therapeutics that are more effective. In this article, we have discussed the vicissitudes that the scientists have gone through to improve prediction accuracy, and examined the effective policies in predicting from different aspects, including the construction of high quality MSA, providing informative input features, and progresses in deep learning approaches. We have also briefly touched upon transitioning from predicting single-chain protein structures to predicting protein complex structures. Our findings point towards promoting open research environments to support the objectives of protein structure prediction.

Predicting protein conformational motions using energetic frustration analysis and AlphaFold2

Proteins perform their biological functions through motion. Although high throughput prediction of the three-dimensional static structures of proteins has proved feasible using deep-learning-based methods, predicting the conformational motions remains a challenge. Purely data-driven machine learning methods encounter difficulty for addressing such motions because available laboratory data on conformational motions are still limited. In this work, we develop a method for generating protein allosteric motions by integrating physical energy landscape information into deep-learning-based methods. We show that local energetic frustration, which represents a quantification of the local features of the energy landscape governing protein allosteric dynamics, can be utilized to empower AlphaFold2 (AF2) to predict protein conformational motions. Starting from ground state static structures, this integrative method generates alternative structures as well as pathways of protein conformational motions, using a progressive enhancement of the energetic frustration features in the input multiple sequence alignment sequences. For a model protein adenylate kinase, we show that the generated conformational motions are consistent with available experimental and molecular dynamics simulation data. Applying the method to another two proteins KaiB and ribose-binding protein, which involve large-amplitude conformational changes, can also successfully generate the alternative conformations. We also show how to extract overall features of the AF2 energy landscape topography, which has been considered by many to be black box. Incorporating physical knowledge into deep-learning-based structure prediction algorithms provides a useful strategy to address the challenges of dynamic structure prediction of allosteric proteins.

AlphaFold predictions of fold-switched conformations are driven by structure memorization

Recent work suggests that AlphaFold (AF)-a deep learning-based model that can accurately infer protein structure from sequence-may discern important features of folded protein energy landscapes, defined by the diversity and frequency of different conformations in the folded state. Here, we test the limits of its predictive power on fold-switching proteins, which assume two structures with regions of distinct secondary and/or tertiary structure. We find that (1) AF is a weak predictor of fold switching and (2) some of its successes result from memorization of training-set structures rather than learned protein energetics. Combining >280,000 models from several implementations of AF2 and AF3, a 35% success rate was achieved for fold switchers likely in AF's training sets. AF2's confidence metrics selected against models consistent with experimentally determined fold-switching structures and failed to discriminate between low and high energy conformations. Further, AF captured only one out of seven experimentally confirmed fold switchers outside of its training sets despite extensive sampling of an additional ~280,000 models. Several observations indicate that AF2 has memorized structural information during training, and AF3 misassigns coevolutionary restraints. These limitations constrain the scope of successful predictions, highlighting the need for physically based methods that readily predict multiple protein conformations.

CDCA7 is an evolutionarily conserved hemimethylated DNA sensor in eukaryotes

Mutations of the SNF2 family ATPase HELLS and its activator CDCA7 cause immunodeficiency, centromeric instability, and facial anomalies syndrome, characterized by DNA hypomethylation at heterochromatin. It remains unclear why CDCA7-HELLS is the sole nucleosome remodeling complex whose deficiency abrogates the maintenance of DNA methylation. We here identify the unique zinc-finger domain of CDCA7 as an evolutionarily conserved hemimethylation-sensing zinc finger (HMZF) domain. Cryo-electron microscopy structural analysis of the CDCA7-nucleosome complex reveals that the HMZF domain can recognize hemimethylated CpG in the outward-facing DNA major groove within the nucleosome core particle, whereas UHRF1, the critical activator of the maintenance methyltransferase DNMT1, cannot. CDCA7 recruits HELLS to hemimethylated chromatin and facilitates UHRF1-mediated H3 ubiquitylation associated with replication-uncoupled maintenance DNA methylation. We propose that the CDCA7-HELLS nucleosome remodeling complex assists the maintenance of DNA methylation on chromatin by sensing hemimethylated CpG that is otherwise inaccessible to UHRF1 and DNMT1.

Development and tissue specific expression of RAPGEF1 (C3G) transcripts having exons encoding disordered segments with predicted regulatory function

BACKGROUND

The ubiquitously expressed Guanine nucleotide exchange factor, RAPGEF1 (C3G), is essential for early development of mouse embryos. It functions to regulate gene expression and cytoskeletal reorganization, thereby controlling cell proliferation and differentiation. While multiple transcripts have been predicted, their expression in mouse tissues has not been investigated in detail.

METHODS & RESULTS

Full length RAPGEF1 isoforms primarily arise due to splicing at two hotspots, one involving exon-3, and the other involving exons 12-14 incorporating amino acids immediately following the Crk binding region of the protein. These isoforms vary in expression across embryonic and adult organs. We detected the presence of unannotated, and unpredicted transcripts with incorporation of cassette exons in various combinations, specifically in the heart, brain, testis and skeletal muscle. Isoform switching was detected as myocytes in culture and mouse embryonic stem cells were differentiated to form myotubes, and embryoid bodies respectively. The cassette exons encode a serine-rich polypeptide chain, which is intrinsically disordered, and undergoes phosphorylation. In silico structural analysis using AlphaFold indicated that the presence of cassette exons alters intra-molecular interactions, important for regulating catalytic activity. LZerD based docking studies predicted that the isoforms with one or more cassette exons differ in interaction with their target GTPase, RAP1A.

CONCLUSIONS

Our results demonstrate the expression of novel RAPGEF1 isoforms, and predict cassette exon inclusion as an additional means of regulating RAPGEF1 activity in various tissues and during differentiation.

MS2/GmAMS1 encodes a bHLH transcription factor important for tapetum degeneration in soybean

KEY MESSAGE

GmAMS1 is the only functional AMS and works with GmTDF1-1 and GmMS3 to orchestrate the tapetum degeneration in soybean. Heterosis could significantly increase the production of major crops as well as soybean [Glycine max (L.) Merr.]. Stable male-sterile/female-fertile mutants including ms2 are useful resources to apply in soybean hybrid production. Here, we identified the detailed mutated sites of two classic mutants ms2 (Eldorado) and ms2 (Ames) in MS2/GmAMS1 via the high-throughput sequencing method. Subsequently, we verified that GmAMS1, a bHLH transcription factor, is the only functional AMS member in soybean through the complementary experiment in Arabidopsis; and elucidated the dysfunction of its homolog GmAMS2 is caused by the premature stop codon in the gene's coding sequence. Further qRT-PCR analysis and protein-protein interaction assays indicated GmAMS1 is required for expressing downstream members in the putative DYT1-TDF1-AMS-MYB80/MYB103/MS188-MS1 cascade module, and might regulate the upstream members in a feedback mechanism. GmAMS1 could interact with GmTDF1-1 and GmMS3 via different region, which contributes to dissect the mechanism in the tapetum degeneration process. Additionally, as a core member in the conserved cascade module controlling the tapetum development and degeneration, AMS is conservatively present in all land plant lineages, implying that AMS-mediated signaling pathway has been established before land plants diverged, which provides further insight into the tapetal evolution.

Computational structural prediction and chemical inhibition of the human mitochondrial pyruvate carrier protein heterodimer complex

The mitochondrial pyruvate carrier (MPC) plays a role in numerous diseases including neurodegeneration, metabolically dependent cancers, and the development of insulin resistance. Several previous studies in genetic mouse models or with existing inhibitors suggest that inhibition of the MPC could be used as a viable therapeutic strategy in these diseases. However, the MPC's structure is unknown, making it difficult to screen for and develop therapeutically viable inhibitors. Currently known MPC inhibitors would make for poor drugs due to their poor pharmacokinetic properties, or in the case of the thiazolidinediones (TZDs), off-target specificity for peroxisome-proliferator activated receptor gamma (PPARγ) leads to unwanted side effects. In this study, we develop several structural models for the MPC heterodimer complex and investigate the chemical interactions required for the binding of these known inhibitors to MPC and PPARγ. Based on these models, the MPC most likely takes on outward-facing (OF) and inward-facing (IF) conformations during pyruvate transport, and inhibitors likely plug the carrier to inhibit pyruvate transport. Although some chemical interactions are similar between MPC and PPARγ binding, there is likely enough difference to reduce PPARγ specificity for future development of novel, more specific MPC inhibitors.

ARID3C Acts as a Regulator of Monocyte-to-Macrophage Differentiation Interacting with NPM1

ARID3C is a protein located on human chromosome 9 and expressed at low levels in various organs, yet its biological function has not been elucidated. In this study, we investigated both the cellular localization and function of ARID3C. Employing a combination of LC-MS/MS and deep learning techniques, we identified NPM1 as a binding partner for ARID3C's nuclear shuttling. ARID3C was found to predominantly localize with the nucleus, where it functioned as a transcription factor for genes STAT3, STAT1, and JUNB, thereby facilitating monocyte-to-macrophage differentiation. The precise binding sites between ARID3C and NPM1 were predicted by AlphaFold2. Mutating this binding site prevented ARID3C from interacting with NPM1, resulting in its retention in the cytoplasm instead of translocation to the nucleus. Consequently, ARID3C lost its ability to bind to the promoters of target genes, leading to a loss of monocyte-to-macrophage differentiation. Collectively, our findings indicate that ARID3C forms a complex with NPM1 to translocate to the nucleus, acting as a transcription factor that promotes the expression of the genes involved in monocyte-to-macrophage differentiation.

MAGPIE: An interactive tool for visualizing and analyzing protein-ligand interactions

Quantitative tools to compile and analyze biomolecular interactions among chemically diverse binding partners would improve therapeutic design and aid in studying molecular evolution. Here we present Mapping Areas of Genetic Parsimony In Epitopes (MAGPIE), a publicly available software package for simultaneously visualizing and analyzing thousands of interactions between a single protein or small molecule ligand (the "target") and all of its protein binding partners ("binders"). MAGPIE generates an interactive three-dimensional visualization from a set of protein complex structures that share the target ligand, as well as sequence logo-style amino acid frequency graphs that show all the amino acids from the set of protein binders that interact with user-defined target ligand positions or chemical groups. MAGPIE highlights all the salt bridge and hydrogen bond interactions made by the target in the visualization and as separate amino acid frequency graphs. Finally, MAGPIE collates the most common target-binder interactions as a list of "hotspots," which can be used to analyze trends or guide the de novo design of protein binders. As an example of the utility of the program, we used MAGPIE to probe how different antibody fragments bind a viral antigen; how a common metabolite binds diverse protein partners; and how two ligands bind orthologs of a well-conserved glycolytic enzyme for a detailed understanding of evolutionarily conserved interactions involved in its activation and inhibition. MAGPIE is implemented in Python 3 and freely available at https://github.com/glasgowlab/MAGPIE, along with sample datasets, usage examples, and helper scripts to prepare input structures.

Integrating Large-Scale Protein Structure Prediction into Human Genetics Research

The last five years have seen impressive progress in deep learning models applied to protein research. Most notably, sequence-based structure predictions have seen transformative gains in the form of AlphaFold2 and related approaches. Millions of missense protein variants in the human population lack annotations, and these computational methods are a valuable means to prioritize variants for further analysis. Here, we review the recent progress in deep learning models applied to the prediction of protein structure and protein variants, with particular emphasis on their implications for human genetics and health. Improved prediction of protein structures facilitates annotations of the impact of variants on protein stability, protein-protein interaction interfaces, and small-molecule binding pockets. Moreover, it contributes to the study of host-pathogen interactions and the characterization of protein function. As genome sequencing in large cohorts becomes increasingly prevalent, we believe that better integration of state-of-the-art protein informatics technologies into human genetics research is of paramount importance.

Chemical cross-linking and mass spectrometry enabled systems-level structural biology

Structural information on protein-protein interactions (PPIs) is essential for improved understanding of regulatory interactome networks that confer various physiological and pathological responses. Additionally, maladaptive PPIs constitute desirable therapeutic targets due to inherently high disease state specificity. Recent advances in chemical cross-linking strategies coupled with mass spectrometry (XL-MS) have positioned XL-MS as a promising technology to not only elucidate the molecular architecture of individual protein assemblies, but also to characterize proteome-wide PPI networks. Moreover, quantitative in vivo XL-MS provides a new capability for the visualization of cellular interactome dynamics elicited by drug treatments, disease states, or aging effects. The emerging field of XL-MS based complexomics enables unique insights on protein moonlighting and protein complex remodeling. These techniques provide complimentary information necessary for in-depth structural interactome studies to better comprehend how PPIs mediate function in living systems.

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.

PPTC7 antagonizes mitophagy by promoting BNIP3 and NIX degradation via SCFFBXL4

Mitophagy must be carefully regulated to ensure that cells maintain appropriate numbers of functional mitochondria. The SCFFBXL4 ubiquitin ligase complex suppresses mitophagy by controlling the degradation of BNIP3 and NIX mitophagy receptors, and FBXL4 mutations result in mitochondrial disease as a consequence of elevated mitophagy. Here, we reveal that the mitochondrial phosphatase PPTC7 is an essential cofactor for SCFFBXL4-mediated destruction of BNIP3 and NIX, suppressing both steady-state and induced mitophagy. Disruption of the phosphatase activity of PPTC7 does not influence BNIP3 and NIX turnover. Rather, a pool of PPTC7 on the mitochondrial outer membrane acts as an adaptor linking BNIP3 and NIX to FBXL4, facilitating the turnover of these mitophagy receptors. PPTC7 accumulates on the outer mitochondrial membrane in response to mitophagy induction or the absence of FBXL4, suggesting a homoeostatic feedback mechanism that attenuates high levels of mitophagy. We mapped critical residues required for PPTC7-BNIP3/NIX and PPTC7-FBXL4 interactions and their disruption interferes with both BNIP3/NIX degradation and mitophagy suppression. Collectively, these findings delineate a complex regulatory mechanism that restricts BNIP3/NIX-induced mitophagy.

The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins

Artificial intelligence-driven advances in protein structure prediction in recent years have raised the question: has the protein structure-prediction problem been solved? Here, with a focus on nonglobular proteins, we highlight the many strengths and potential weaknesses of DeepMind's AlphaFold2 in the context of its biological and therapeutic applications. We summarize the subtleties associated with evaluation of AlphaFold2 model quality and reliability using the predicted local distance difference test (pLDDT) and predicted aligned error (PAE) values. We highlight various classes of proteins that AlphaFold2 can be applied to and the caveats involved. Concrete examples of how AlphaFold2 models can be integrated with experimental data in the form of small-angle X-ray scattering (SAXS), solution NMR, cryo-electron microscopy (cryo-EM) and X-ray diffraction are discussed. Finally, we highlight the need to move beyond structure prediction of rigid, static structural snapshots toward conformational ensembles and alternate biologically relevant states. The overarching theme is that careful consideration is due when using AlphaFold2-generated models to generate testable hypotheses and structural models, rather than treating predicted models as de facto ground truth structures.

Structures, dynamics, complexes, and functions: From classic computation to artificial intelligence

Computational approaches can provide highly detailed insight into the molecular recognition processes that underlie drug binding, the assembly of protein complexes, and the regulation of biological functional processes. Classical simulation methods can bridge a wide range of length- and time-scales typically involved in such processes. Lately, automated learning and artificial intelligence methods have shown the potential to expand the reach of physics-based approaches, ushering in the possibility to model and even design complex protein architectures. The synergy between atomistic simulations and AI methods is an emerging frontier with a huge potential for advances in structural biology. Herein, we explore various examples and frameworks for these approaches, providing select instances and applications that illustrate their impact on fundamental biomolecular problems.

Integrative modeling meets deep learning: Recent advances in modeling protein assemblies

Recent progress in protein structure prediction based on deep learning revolutionized the field of Structural Biology. Beyond single proteins, it also enabled high-throughput prediction of structures of protein-protein interactions. Despite the success in predicting complex structures, large macromolecular assemblies still require specialized approaches. Here we describe recent advances in modeling macromolecular assemblies using integrative and hierarchical approaches. We highlight applications that predict protein-protein interactions and challenges in modeling complexes based on the interaction networks, including the prediction of complex stoichiometry and heterogeneity.