{"title":"Dynamics-based drug discovery by time-resolved cryo-EM","authors":"Youdong Mao","doi":"10.1016/j.sbi.2025.103001","DOIUrl":"10.1016/j.sbi.2025.103001","url":null,"abstract":"<div><div>Rational structure-based drug design (SBDD) depends on high-resolution structural models of target macromolecules or their complexes. However, the lack of atomic-level functional molecular dynamics hinders the applications of SBDD and limits their effective translation into clinically successful therapeutics. Time-resolved cryo-electron microscopy (cryo-EM) has emerged as a powerful tool in structural biology, capable of capturing high-resolution snapshots of biomolecular machines in action. Unlike molecular dynamics (MD) simulations, time-resolved cryo-EM can visualize rare intermediate states across a broader range of timescales, providing invaluable insights into drug-binding kinetics, dynamic protein-ligand interactions, and allosteric regulation. Integration of time-resolved cryo-EM with machine learning (ML) and artificial intelligence (AI) expands SBDD into a dynamics-based approach, allowing for more accurate pharmacological modeling of challenging drug targets that are beyond the reach of MD simulations. Time-resolved cryo-EM can help researchers to identify novel druggable conformations, overcome drug resistance, and reduce the time and cost of clinical translations. Despite current challenges, the future development of time-resolved cryo-EM with AI and <em>in situ</em> imaging strategy, such as cryo-electron tomography, holds the potential to revolutionize drug discovery by revealing <em>in vivo</em> molecular dynamics of drug actions at an unprecedented spatiotemporal scale.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103001"},"PeriodicalIF":6.1,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143455034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Atabey Ünlü , Erva Ulusoy , Melih Gökay Yiğit , Melih Darcan , Tunca Doğan
{"title":"Protein language models for predicting drug–target interactions: Novel approaches, emerging methods, and future directions","authors":"Atabey Ünlü , Erva Ulusoy , Melih Gökay Yiğit , Melih Darcan , Tunca Doğan","doi":"10.1016/j.sbi.2025.103017","DOIUrl":"10.1016/j.sbi.2025.103017","url":null,"abstract":"<div><div>Identifying new drug candidates remains a critical and complex challenge in drug development. Recent advances in deep learning have demonstrated significant potential to accelerate this process, particularly through the use of protein language models (pLMs). These models aim to effectively capture the structural and functional properties of proteins by embedding them in high-dimensional spaces, thereby providing powerful tools for predictive tasks. This review examines the application of pLMs in drug-target interaction (DTI) prediction, addressing both small-molecule and protein-based therapeutics. We explore diverse methodologies, including end-to-end learning models and those that leverage pre-trained foundational pLMs. Furthermore, we highlight the role of heterogeneous data integration—ranging from protein structures to knowledge graphs—to improve the accuracy of DTI predictions. Despite notable progress, challenges persist in accurately identifying DTIs, mainly due to data-related limitations and algorithmic constraints. Future research directions include utilising multimodal learning approaches, incorporating temporal/dynamic interaction data into training, and employing novel deep learning architectures to refine protein representations, gain a deeper understanding of biological context regarding molecular interactions, and, thus, advance the DTI prediction field.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103017"},"PeriodicalIF":6.1,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143455033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Teaching AI to speak protein","authors":"Michael Heinzinger , Burkhard Rost","doi":"10.1016/j.sbi.2025.102986","DOIUrl":"10.1016/j.sbi.2025.102986","url":null,"abstract":"<div><div>Large Language Models for proteins, namely protein Language Models (<u>pLMs</u>), have begun to provide an important alternative to capturing the information encoded in a protein sequence in computers. Arguably, pLMs have advanced importantly to understanding aspects of the <em>language of life</em> as written in proteins, and through this understanding, they are becoming an increasingly powerful means of advancing protein prediction, e.g., in the prediction of molecular function as expressed by identifying binding residues or variant effects. While benefitting from the same technology, protein structure prediction remains one of the few applications for which only using pLM embeddings from single sequences appears not to improve over or match the state-of-the-art. Fine-tuning foundation pLMs enhances efficiency and accuracy of solutions, in particular in cases with few experimental annotations. pLMs facilitate the integration of computational and experimental biology, of AI and wet-lab, in particular toward a new era of protein design.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 102986"},"PeriodicalIF":6.1,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143455035","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Prediction of structural variation","authors":"Yogesh Kalakoti, Airy Sanjeev, Björn Wallner","doi":"10.1016/j.sbi.2025.103003","DOIUrl":"10.1016/j.sbi.2025.103003","url":null,"abstract":"<div><div>Proteins are dynamic molecules that transition between conformational states to perform their functions, and characterizing the protein ensemble is important for understanding biology and therapeutic applications. While recent breakthroughs in machine learning have enabled the prediction of high-quality static models of individual proteins, generating reliable estimates of their conformational ensembles remains a challenge. Several recent methods have tried to utilize the evolutionary and structural features captured by effective sequence-to-structure models to enhance conformational diversity in generated models. Most of these approaches involve adapting existing inference pipelines, such as AlphaFold 2, combined with sampling techniques to induce the generation of diverse conformational states. Here, we describe the general problem of predicting structural variations in protein systems, explain the methods designed to address this challenge, explore why they are effective, discuss their limitations, and suggest potential future directions.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103003"},"PeriodicalIF":6.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143455037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Chromatin domains in the cell: Phase separation and condensation","authors":"Shin Fujishiro , Masaki Sasai , Kazuhiro Maeshima","doi":"10.1016/j.sbi.2025.103006","DOIUrl":"10.1016/j.sbi.2025.103006","url":null,"abstract":"<div><div>Negatively charged genomic DNA wraps around positively charged core histone octamers to form nucleosomes, which, along with proteins and RNAs, self-organize into chromatin within the nucleus. In eukaryotic cells, chromatin forms loops that collapse into chromatin domains and serve as functional units of the genome. Chromatin domains vary in physical properties based on gene activity and are assembled into A (euchromatin) and B (heterochromatin) compartments. Since various factors—such as chromatin-binding proteins, histone modifications, transcriptional states, depletion attraction, and cations—can significantly impact chromatin organization, the formation processes of these hierarchical structures remain unclear. No single imaging, genomics, or modeling method can provide a complete picture of the process. Beautiful models can sometimes fool our thinking. In this short review, we critically discuss the formation mechanisms of the chromatin domain in the cell from a physical point of view, including phase separation and condensation.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103006"},"PeriodicalIF":6.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143445627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chentong Wang , Sarah Alamdari , Carles Domingo-Enrich , Ava P. Amini , Kevin K. Yang
{"title":"Toward deep learning sequence–structure co-generation for protein design","authors":"Chentong Wang , Sarah Alamdari , Carles Domingo-Enrich , Ava P. Amini , Kevin K. Yang","doi":"10.1016/j.sbi.2025.103018","DOIUrl":"10.1016/j.sbi.2025.103018","url":null,"abstract":"<div><div>Deep generative models that learn from the distribution of natural protein sequences and structures may enable the design of new proteins with valuable functions. While the majority of today’s models focus on generating either sequences or structures, emerging co-generation methods promise more accurate and controllable protein design, ideally achieved by modeling both modalities simultaneously. Here we review recent advances in deep generative models for protein design, with a particular focus on sequence-structure co-generation methods. We describe the key methodological and evaluation principles underlying these methods, highlight recent advances from the literature, and discuss opportunities for continued development of sequence-structure co-generation approaches.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103018"},"PeriodicalIF":6.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143455036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Andreas Bjerregaard , Peter Mørch Groth , Søren Hauberg , Anders Krogh , Wouter Boomsma
{"title":"Foundation models of protein sequences: A brief overview","authors":"Andreas Bjerregaard , Peter Mørch Groth , Søren Hauberg , Anders Krogh , Wouter Boomsma","doi":"10.1016/j.sbi.2025.103004","DOIUrl":"10.1016/j.sbi.2025.103004","url":null,"abstract":"<div><div>Protein sequence models have evolved from simple statistics of aligned families to versatile foundation models of evolutionary scale. Enabled by self-supervised learning and an abundance of protein sequence data, such foundation models now play a central role in protein science. They facilitate rich representations, powerful generative design, and fine-tuning across diverse domains. In this review, we trace modeling developments and categorize them into methodological trends over the modalities they describe and the contexts they condition upon. Following a brief historical overview, we focus our attention on the most recent trends and outline future perspectives.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103004"},"PeriodicalIF":6.1,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143445626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"From part to whole: AI-driven progress in fragment-based drug discovery","authors":"Jinhyeok Yoo , Wonkyeong Jang , Woong-Hee Shin","doi":"10.1016/j.sbi.2025.102995","DOIUrl":"10.1016/j.sbi.2025.102995","url":null,"abstract":"<div><div>Fragment-based drug discovery is a technique that finds potent binding fragments to the binding hotspots and makes them a hit compound. The combination of fragments allows us to explore the large chemical space. Thus, it becomes an effective methodology for identifying lead compounds. Three concepts have been introduced to make the fragments into the compound: growing, merging, and linking. Recently, growing and merging techniques using AI have significantly improved the accuracy and efficiency of molecular design. In this review, recent techniques such as VAE, reinforcement learning, and SE(3)-equivariant models will be discussed. These methods enable precise molecular structure exploration and optimization. Additionally, we address techniques utilizing diffusion models, language models, and deep evolutionary learning. We also introduce linker optimization methods using reinforcement learning and deep learning-based models. This progress of fragment-based drug discovery methods with AI opens the possibility of discovering the vast chemical space with high efficiency.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 102995"},"PeriodicalIF":6.1,"publicationDate":"2025-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143429491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Combining cryo-electron microscopy (cryo-EM) with orthogonal solution state methods to define the molecular basis of the phosphoprotein phosphatase family regulation and substrate specificity","authors":"Wolfgang Peti , Sathish K.R. Padi , Rebecca Page","doi":"10.1016/j.sbi.2025.102992","DOIUrl":"10.1016/j.sbi.2025.102992","url":null,"abstract":"<div><div>Protein phosphatases are dynamic enzymes that exhibit complex regulatory mechanisms, with disruptions in these regulatory processes associated with disease. It is now clear that many phosphatases assemble into large macromolecular complexes via the interaction of phosphatase-specific regulatory proteins and substrates containing short linear motifs (SLiMs) or short helical motifs (SHelMs). Here, we review how cryo-electron microscopy (cryo-EM) integrated with orthogonal methods to study dynamic protein–protein interactions (NMR spectroscopy, hydrogen-deuterium exchange mass spectrometry, among others) is leading to new discoveries about the mechanisms controlling phosphatase assembly, substrate recruitment and dephosphorylation and, in turn, are providing novel strategies for targeting phosphatase-related diseases. This review focuses on the recently determined structures and regulation of the phosphoprotein phosphatase (PPP) family of ser/thr phosphatases—PP1, PP2A, Calcineurin and PP5.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 102992"},"PeriodicalIF":6.1,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Bias in, bias out – AlphaFold-Multimer and the structural complexity of protein interfaces","authors":"Joelle Morgan Strom, Katja Luck","doi":"10.1016/j.sbi.2025.103002","DOIUrl":"10.1016/j.sbi.2025.103002","url":null,"abstract":"<div><div>A structural understanding of protein–protein interactions is a key component of many facets of applied molecular biology research. AlphaFold-Multimer (AF-MM) provided a breakthrough in the ability to predict protein–protein interface structure. However, the available training data for this model and the resulting benchmarking and validation efforts show a bias toward interactions between more ordered regions of proteins. Here we highlight some of the successes and limitations of AF-MM and discuss available methods and future directions to enable balanced prediction of all interface types.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 103002"},"PeriodicalIF":6.1,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143388119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}