Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2025.102990
Arjun Dosajh , Prakul Agrawal , Prathit Chatterjee, U. Deva Priyakumar
Recent progress and development of artificial intelligence and machine learning (AI/ML) techniques have enabled addressing complex biomolecular problems. AI/ML models learn the underlying distribution of data they are trained on and when exposed to new inputs, they make predictions based on patterns and relationships previously observed in the training set. Further, generative artificial intelligence (GenAI) can be used to accurately generate protein structure or sequence from specific selected properties. This review specifically focuses on the applications of AI/ML in predicting important functional properties of proteins, and the potential prospects of reverse-engineering in depicting the sequence and structure, from available protein-property information.
{"title":"Modern machine learning methods for protein property prediction","authors":"Arjun Dosajh , Prakul Agrawal , Prathit Chatterjee, U. Deva Priyakumar","doi":"10.1016/j.sbi.2025.102990","DOIUrl":"10.1016/j.sbi.2025.102990","url":null,"abstract":"<div><div>Recent progress and development of artificial intelligence and machine learning (AI/ML) techniques have enabled addressing complex biomolecular problems. AI/ML models learn the underlying distribution of data they are trained on and when exposed to new inputs, they make predictions based on patterns and relationships previously observed in the training set. Further, generative artificial intelligence (GenAI) can be used to accurately generate protein structure or sequence from specific selected properties. This review specifically focuses on the applications of AI/ML in predicting important functional properties of proteins, and the potential prospects of reverse-engineering in depicting the sequence and structure, from available protein-property information.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102990"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143064149","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2024.102958
Thibault Orand, Malene Ringkjøbing Jensen
Advances in the characterization of intrinsically disordered proteins (IDPs) have unveiled a remarkably complex and diverse interaction landscape, including coupled folding and binding, highly dynamic complexes, multivalent interactions, and even interactions between entirely disordered proteins. Here we review recent examples of IDP binding mechanisms elucidated by experimental techniques such as nuclear magnetic resonance spectroscopy, single-molecule Förster resonance energy transfer, and stopped-flow fluorescence. These techniques provide insights into the structural details of transition pathways and complex intermediates, and they capture the dynamics of IDPs within complexes. Furthermore, we discuss the growing role of artificial intelligence, exemplified by AlphaFold, in identifying interaction sites within IDPs and predicting their bound-state structures. Our review highlights the powerful complementarity between experimental methods and artificial intelligence-based approaches in advancing our understanding of the intricate interaction landscape of IDPs.
{"title":"Binding mechanisms of intrinsically disordered proteins: Insights from experimental studies and structural predictions","authors":"Thibault Orand, Malene Ringkjøbing Jensen","doi":"10.1016/j.sbi.2024.102958","DOIUrl":"10.1016/j.sbi.2024.102958","url":null,"abstract":"<div><div>Advances in the characterization of intrinsically disordered proteins (IDPs) have unveiled a remarkably complex and diverse interaction landscape, including coupled folding and binding, highly dynamic complexes, multivalent interactions, and even interactions between entirely disordered proteins. Here we review recent examples of IDP binding mechanisms elucidated by experimental techniques such as nuclear magnetic resonance spectroscopy, single-molecule Förster resonance energy transfer, and stopped-flow fluorescence. These techniques provide insights into the structural details of transition pathways and complex intermediates, and they capture the dynamics of IDPs within complexes. Furthermore, we discuss the growing role of artificial intelligence, exemplified by AlphaFold, in identifying interaction sites within IDPs and predicting their bound-state structures. Our review highlights the powerful complementarity between experimental methods and artificial intelligence-based approaches in advancing our understanding of the intricate interaction landscape of IDPs.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102958"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142909281","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2024.102957
Junji Iwahara, David C. Williams Jr.
{"title":"Editorial overview: New perspectives on the structure and dynamics of protein-nucleic acid interactions","authors":"Junji Iwahara, David C. Williams Jr.","doi":"10.1016/j.sbi.2024.102957","DOIUrl":"10.1016/j.sbi.2024.102957","url":null,"abstract":"","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102957"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142779643","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2025.102983
Alexander Greenshields-Watson , Odysseas Vavourakis , Fabian C. Spoendlin , Matteo Cagiada , Charlotte M. Deane
Therapeutic antibodies are manufactured, stored and administered in the free state; this makes understanding the unbound form key to designing and improving development pipelines. Prediction of unbound antibodies is challenging, specifically modelling of the CDRH3 loop, where inaccuracies are potentially worse due to a bias in structural data towards antibody-antigen complexes. This class imbalance provides a challenge for deep learning models trained on this data, potentially limiting generalisation to unbound forms.
Here we discuss the importance of unbound structures in antibody development pipelines. We explore how the latest generation of structure predictors can provide new insights and assess how conformational heterogeneity may influence binding kinetics. We hypothesise that generative models may address some of these issues. While prediction of antibodies in complex is essential, we should not ignore the need for progress in modelling the unbound form.
{"title":"Challenges and compromises: Predicting unbound antibody structures with deep learning","authors":"Alexander Greenshields-Watson , Odysseas Vavourakis , Fabian C. Spoendlin , Matteo Cagiada , Charlotte M. Deane","doi":"10.1016/j.sbi.2025.102983","DOIUrl":"10.1016/j.sbi.2025.102983","url":null,"abstract":"<div><div>Therapeutic antibodies are manufactured, stored and administered in the free state; this makes understanding the unbound form key to designing and improving development pipelines. Prediction of unbound antibodies is challenging, specifically modelling of the CDRH3 loop, where inaccuracies are potentially worse due to a bias in structural data towards antibody-antigen complexes. This class imbalance provides a challenge for deep learning models trained on this data, potentially limiting generalisation to unbound forms.</div><div>Here we discuss the importance of unbound structures in antibody development pipelines. We explore how the latest generation of structure predictors can provide new insights and assess how conformational heterogeneity may influence binding kinetics. We hypothesise that generative models may address some of these issues. While prediction of antibodies in complex is essential, we should not ignore the need for progress in modelling the unbound form.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102983"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143037416","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2025.102984
Mesih Kilinc , Kejue Jia , Robert L. Jernigan
There is an ever-increasing need for accurate and efficient methods to identify protein homologs. Traditionally, sequence similarity-based methods have dominated protein homolog identification for function identification, but these struggle when the sequence identity between the pairs is low. Recently, transformer architecture-based deep learning methods have achieved breakthrough performances in many fields. One type of model that uses transformer architecture is the protein language model (pLM). Here, we describe methods that use pLMs for protein homolog identification intended for function identification and describe their strengths and weaknesses. Several important ideas emerge, such as filtering the substitution matrix generated from embeddings, selecting specific pLM layers for specific purposes, compressing the embeddings, and dividing proteins into domains before searching for homologs that improve remote homolog detection accuracy considerably. All of these approaches produce huge numbers of new homologs that can reliably extend the reach of protein relationships for a deeper understanding of evolution and many other problems.
{"title":"Major advances in protein function assignment by remote homolog detection with protein language models – A review","authors":"Mesih Kilinc , Kejue Jia , Robert L. Jernigan","doi":"10.1016/j.sbi.2025.102984","DOIUrl":"10.1016/j.sbi.2025.102984","url":null,"abstract":"<div><div>There is an ever-increasing need for accurate and efficient methods to identify protein homologs. Traditionally, sequence similarity-based methods have dominated protein homolog identification for function identification, but these struggle when the sequence identity between the pairs is low. Recently, transformer architecture-based deep learning methods have achieved breakthrough performances in many fields. One type of model that uses transformer architecture is the protein language model (pLM). Here, we describe methods that use pLMs for protein homolog identification intended for function identification and describe their strengths and weaknesses. Several important ideas emerge, such as filtering the substitution matrix generated from embeddings, selecting specific pLM layers for specific purposes, compressing the embeddings, and dividing proteins into domains before searching for homologs that improve remote homolog detection accuracy considerably. All of these approaches produce huge numbers of new homologs that can reliably extend the reach of protein relationships for a deeper understanding of evolution and many other problems.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102984"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143045750","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2024.102956
Eric Conway, Daniel R. Larson
{"title":"Editorial overview: 3D Genome Chromatin organization and regulation","authors":"Eric Conway, Daniel R. Larson","doi":"10.1016/j.sbi.2024.102956","DOIUrl":"10.1016/j.sbi.2024.102956","url":null,"abstract":"","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102956"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142779641","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}
Cryo-electron microscopy (Cryo-EM) has revolutionized structural biology by enabling the determination of macromolecular structures that were challenging to study with conventional methods. Processing cryo-EM data involves several computational steps to derive three-dimensional structures from raw projections. Recent advancements in artificial intelligence (AI) including deep learning have significantly improved the performance of these processes. In this review, we discuss state-of-the-art AI-based techniques used in key steps of cryo-EM data processing, including macromolecular structure modeling and heterogeneity analysis.
{"title":"AI-based methods for biomolecular structure modeling for Cryo-EM","authors":"Farhanaz Farheen , Genki Terashi , Han Zhu , Daisuke Kihara","doi":"10.1016/j.sbi.2025.102989","DOIUrl":"10.1016/j.sbi.2025.102989","url":null,"abstract":"<div><div>Cryo-electron microscopy (Cryo-EM) has revolutionized structural biology by enabling the determination of macromolecular structures that were challenging to study with conventional methods. Processing cryo-EM data involves several computational steps to derive three-dimensional structures from raw projections. Recent advancements in artificial intelligence (AI) including deep learning have significantly improved the performance of these processes. In this review, we discuss state-of-the-art AI-based techniques used in key steps of cryo-EM data processing, including macromolecular structure modeling and heterogeneity analysis.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102989"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143045837","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2024.102976
Alexander I.M. Sever , Rashik Ahmed , Philip Rößler , Lewis E. Kay
The tools of structural biology have undergone remarkable advances in the past decade. These include new computational and experimental approaches that have enabled studies at a level of detail – and ease – that were not previously possible. Yet, significant deficiencies in our understanding of biomolecular function remain and new challenges must be overcome to go beyond static pictures towards a description of function in terms of structural dynamics. Solution Nuclear Magnetic Resonance (NMR) spectroscopy has emerged as a powerful technique for atomic resolution studies of the dynamics of a wide range of biomolecules, including molecular machines and the components of phase-separated condensates. Here we highlight some of the very recent advances in these areas that have been driven by NMR.
{"title":"Solution NMR goes big: Atomic resolution studies of protein components of molecular machines and phase-separated condensates","authors":"Alexander I.M. Sever , Rashik Ahmed , Philip Rößler , Lewis E. Kay","doi":"10.1016/j.sbi.2024.102976","DOIUrl":"10.1016/j.sbi.2024.102976","url":null,"abstract":"<div><div>The tools of structural biology have undergone remarkable advances in the past decade. These include new computational and experimental approaches that have enabled studies at a level of detail – and ease – that were not previously possible. Yet, significant deficiencies in our understanding of biomolecular function remain and new challenges must be overcome to go beyond static pictures towards a description of function in terms of structural dynamics. Solution Nuclear Magnetic Resonance (NMR) spectroscopy has emerged as a powerful technique for atomic resolution studies of the dynamics of a wide range of biomolecules, including molecular machines and the components of phase-separated condensates. Here we highlight some of the very recent advances in these areas that have been driven by NMR.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102976"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143001536","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2025.102993
Hannah M. Britt , Carol V. Robinson
As health needs in our society evolve, the field of drug discovery must undergo constant innovation and improvement to identify novel targets and drug candidates. Owing to its ability to simultaneously capture biological interactions and provide in-depth molecular characterisation of the species involved, native mass spectrometry is starting to play an important role in this endeavour. Here, we discuss recent contributions that native mass spectrometry has made to drug discovery including deciphering protein-small molecule interactions, unravelling biochemical pathways, and integrating with complementary structural approaches.
{"title":"Traversing the drug discovery landscape using native mass spectrometry","authors":"Hannah M. Britt , Carol V. Robinson","doi":"10.1016/j.sbi.2025.102993","DOIUrl":"10.1016/j.sbi.2025.102993","url":null,"abstract":"<div><div>As health needs in our society evolve, the field of drug discovery must undergo constant innovation and improvement to identify novel targets and drug candidates. Owing to its ability to simultaneously capture biological interactions and provide in-depth molecular characterisation of the species involved, native mass spectrometry is starting to play an important role in this endeavour. Here, we discuss recent contributions that native mass spectrometry has made to drug discovery including deciphering protein-small molecule interactions, unravelling biochemical pathways, and integrating with complementary structural approaches.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"91 ","pages":"Article 102993"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143078796","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}
Pub Date : 2025-02-01DOI: 10.1016/j.sbi.2024.102975
Martijn P. Bemelmans , Zoe Cournia , Kelly L. Damm-Ganamet , Francesco L. Gervasio , Vineet Pande
A number of promising therapeutic target proteins have been considered “undruggable” due to the lack of well-defined ligandable pockets. Substantial research in protein dynamics has elucidated the existence of “cryptic” pockets that only exist transiently and become favorable for binding in the presence of a ligand. These pockets provide an avenue to target challenging proteins, inspiring the development of multiple computational methods. This review highlights established cryptic pocket modeling approaches like mixed solvent molecular dynamics and recent applications of enhanced sampling and AI-based methods in therapeutically relevant proteins.
{"title":"Computational advances in discovering cryptic pockets for drug discovery","authors":"Martijn P. Bemelmans , Zoe Cournia , Kelly L. Damm-Ganamet , Francesco L. Gervasio , Vineet Pande","doi":"10.1016/j.sbi.2024.102975","DOIUrl":"10.1016/j.sbi.2024.102975","url":null,"abstract":"<div><div>A number of promising therapeutic target proteins have been considered “undruggable” due to the lack of well-defined ligandable pockets. Substantial research in protein dynamics has elucidated the existence of “cryptic” pockets that only exist transiently and become favorable for binding in the presence of a ligand. These pockets provide an avenue to target challenging proteins, inspiring the development of multiple computational methods. This review highlights established cryptic pocket modeling approaches like mixed solvent molecular dynamics and recent applications of enhanced sampling and AI-based methods in therapeutically relevant proteins.</div></div>","PeriodicalId":10887,"journal":{"name":"Current opinion in structural biology","volume":"90 ","pages":"Article 102975"},"PeriodicalIF":6.1,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142946100","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号