Deep learning for protein complex structure prediction

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.

High-Speed Atomic Force Microscopy for Filming Protein Molecules in Dynamic Action

Structural biology is currently undergoing a transformation into dynamic structural biology, which reveals the dynamic structure of proteins during their functional activity to better elucidate how they function. Among the various approaches in dynamic structural biology, high-speed atomic force microscopy (HS-AFM) is unique in the ability to film individual molecules in dynamic action, although only topographical information is acquirable. This review provides a guide to the use of HS-AFM for biomolecular imaging and showcases several examples, as well as providing information on up-to-date progress in HS-AFM technology. Finally, we discuss the future prospects of HS-AFM in the context of dynamic structural biology in the upcoming era.

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.

Iron-sulfur protein odyssey: exploring their cluster functional versatility and challenging identification

Iron-sulfur (Fe-S) clusters are an essential and ubiquitous class of protein-bound prosthetic centers that are involved in a broad range of biological processes (e.g. respiration, photosynthesis, DNA replication and repair and gene regulation) performing a wide range of functions including electron transfer, enzyme catalysis, and sensing. In a general manner, Fe-S clusters can gain or lose electrons through redox reactions, and are highly sensitive to oxidation, notably by small molecules such as oxygen and nitric oxide. The [2Fe-2S] and [4Fe-4S] clusters, the most common Fe-S cofactors, are typically coordinated by four amino acid side chains from the protein, usually cysteine thiolates, but other residues (e.g. histidine, aspartic acid) can also be found. While diversity in cluster coordination ensures the functional variety of the Fe-S clusters, the lack of conserved motifs makes new Fe-S protein identification challenging especially when the Fe-S cluster is also shared between two proteins as observed in several dimeric transcriptional regulators and in the mitoribosome. Thanks to the recent development of in cellulo, in vitro, and in silico approaches, new Fe-S proteins are still regularly identified, highlighting the functional diversity of this class of proteins. In this review, we will present three main functions of the Fe-S clusters and explain the difficulties encountered to identify Fe-S proteins and methods that have been employed to overcome these issues.

Structural determinants for activation of the Tau kinase CDK5 by the serotonin receptor 5-HT7R

BACKGROUND

Multiple neurodegenerative diseases are induced by the formation and deposition of protein aggregates. In particular, the microtubule-associated protein Tau leads to the development of so-called tauopathies characterized by the aggregation of hyperphosphorylated Tau within neurons. We recently showed that the constitutive activity of the serotonin receptor 7 (5-HT7R) is required for Tau hyperphosphorylation and aggregation through activation of the cyclin-dependent kinase 5 (CDK5). We also demonstrated physical interaction between 5-HT7R and CDK5 at the plasma membrane suggesting that the 5-HT7R/CDK5 complex is an integral part of the signaling network involved in Tau-mediated pathology.

METHODS

Using biochemical, microscopic, molecular biological, computational and AI-based approaches, we investigated structural requirements for the formation of 5-HT7R/CDK5 complex.

RESULTS

We demonstrated that 5-HT7R domains responsible for coupling to Gs proteins are not involved in receptor interaction with CDK5. We also created a structural model of the 5-HT7R/CDK5 complex and refined the interaction interface. The model predicted two conserved phenylalanine residues, F278 and F281, within the third intracellular loop of 5-HT7R to be potentially important for complex formation. While site-directed mutagenesis of these residues did not influence Gs protein-mediated receptor signaling, replacement of both phenylalanines by alanine residues significantly reduced 5-HT7R/CDK5 interaction and receptor-mediated CDK5 activation, leading to reduced Tau hyperphosphorylation and aggregation. Molecular dynamics simulations of 5-HT7R/CDK5 complex for wild-type and receptor mutants confirmed binding interface stability of the initial model.

CONCLUSIONS

Our results provide a structural basis for the development of novel drugs targeting the 5-HT7R/CDK5 interaction interface for the selective treatment of Tau-related disorders, including frontotemporal dementia and Alzheimer's disease.

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.

An Efficient Method for Isolating and Purifying Nuclei from Mice Brain for Single-Molecule Imaging Using High-Speed Atomic Force Microscopy

Nuclear pore complexes (NPCs) on the nuclear membrane surface have a crucial function in controlling the movement of small molecules and macromolecules between the cell nucleus and cytoplasm through their intricate core channel resembling a spiderweb with several layers. Currently, there are few methods available to accurately measure the dynamics of nuclear pores on the nuclear membranes at the nanoscale. The limitation of traditional optical imaging is due to diffraction, which prevents achieving the required resolution for observing a diverse array of organelles and proteins within cells. Super-resolution techniques have effectively addressed this constraint by enabling the observation of subcellular components on the nanoscale. Nevertheless, it is crucial to acknowledge that these methods often need the use of fixed samples. This also raises the question of how closely a static image represents the real intracellular dynamic system. High-speed atomic force microscopy (HS-AFM) is a unique technique used in the field of dynamic structural biology, enabling the study of individual molecules in motion close to their native states. Establishing a reliable and repeatable technique for imaging mammalian tissue at the nanoscale using HS-AFM remains challenging due to inadequate sample preparation. This study presents the rapid strainer microfiltration (RSM) protocol for directly preparing high-quality nuclei from the mouse brain. Subsequently, we promptly utilize HS-AFM real-time imaging and cinematography approaches to record the spatiotemporal of nuclear pore nano-dynamics from the mouse brain.

Recent Advances and Challenges in Protein Structure Prediction

Artificial intelligence has made significant advances in the field of protein structure prediction in recent years. In particular, DeepMind's end-to-end model, AlphaFold2, has demonstrated the capability to predict three-dimensional structures of numerous unknown proteins with accuracy levels comparable to those of experimental methods. This breakthrough has opened up new possibilities for understanding protein structure and function as well as accelerating drug discovery and other applications in the field of biology and medicine. Despite the remarkable achievements of artificial intelligence in the field, there are still some challenges and limitations. In this Review, we discuss the recent progress and some of the challenges in protein structure prediction. These challenges include predicting multidomain protein structures, protein complex structures, multiple conformational states of proteins, and protein folding pathways. Furthermore, we highlight directions in which further improvements can be conducted.

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.

Challenges in bridging the gap between protein structure prediction and functional interpretation

The rapid evolution of protein structure prediction tools has significantly broadened access to protein structural data. Although predicted structure models have the potential to accelerate and impact fundamental and translational research significantly, it is essential to note that they are not validated and cannot be considered the ground truth. Thus, challenges persist, particularly in capturing protein dynamics, predicting multi-chain structures, interpreting protein function, and assessing model quality. Interdisciplinary collaborations are crucial to overcoming these obstacles. Databases like the AlphaFold Protein Structure Database, the ESM Metagenomic Atlas, and initiatives like the 3D-Beacons Network provide FAIR access to these data, enabling their interpretation and application across a broader scientific community. Whilst substantial advancements have been made in protein structure prediction, further progress is required to address the remaining challenges. Developing training materials, nurturing collaborations, and ensuring open data sharing will be paramount in this pursuit. The continued evolution of these tools and methodologies will deepen our understanding of protein function and accelerate disease pathogenesis and drug development discoveries.

Revolutionizing Medicinal Chemistry: The Application of Artificial Intelligence (AI) in Early Drug Discovery

Artificial intelligence (AI) has permeated various sectors, including the pharmaceutical industry and research, where it has been utilized to efficiently identify new chemical entities with desirable properties. The application of AI algorithms to drug discovery presents both remarkable opportunities and challenges. This review article focuses on the transformative role of AI in medicinal chemistry. We delve into the applications of machine learning and deep learning techniques in drug screening and design, discussing their potential to expedite the early drug discovery process. In particular, we provide a comprehensive overview of the use of AI algorithms in predicting protein structures, drug-target interactions, and molecular properties such as drug toxicity. While AI has accelerated the drug discovery process, data quality issues and technological constraints remain challenges. Nonetheless, new relationships and methods have been unveiled, demonstrating AI's expanding potential in predicting and understanding drug interactions and properties. For its full potential to be realized, interdisciplinary collaboration is essential. This review underscores AI's growing influence on the future trajectory of medicinal chemistry and stresses the importance of ongoing synergies between computational and domain experts.