{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T17:48:09Z","timestamp":1755798489820,"version":"3.44.0"},"reference-count":63,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2025,7,7]],"date-time":"2025-07-07T00:00:00Z","timestamp":1751846400000},"content-version":"vor","delay-in-days":6,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,7,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>A key challenge in protein engineering is understanding how mutations affect protein fitness and stability. Most of current state-of-the-art models fine-tune protein structure prediction or protein language models or even pretrain their own. Despite its widespread use within computational workflows, AlphaFold2 exhibits limited sensitivity in assessing the effects of amino acid point mutations on protein structure, thereby constraining its utility in sequence design and protein engineering. In this work, we propose a simple modification of AlphaFold2 inference that improves the model\u2019s capacity to capture the structural impacts of amino acid mutations. We achieve this by discarding the multiple sequence alignment and masking the template in recycling stages. Moreover, we introduce AFToolkit, a framework that leverages the embeddings of the modified AlphaFold2 model and simple adapter models to solve multiple protein engineering tasks. In contrast to other methods, our approach does not require fine-tuning the AlphaFold2 model or pretraining a new model from scratch on large datasets. It also supports handling multiple mutations, insertions, and deletions by directly modifying the input protein sequence. The proposed approach achieves strong performance across established benchmarks in terms of Spearman correlation: $0.68$ on PTMul, $0.60$ on cDNA-indel, and $0.57$ on C380.<\/jats:p>","DOI":"10.1093\/bib\/bbaf324","type":"journal-article","created":{"date-parts":[[2025,7,7]],"date-time":"2025-07-07T09:55:45Z","timestamp":1751882145000},"source":"Crossref","is-referenced-by-count":0,"title":["AFToolkit: a framework for molecular modeling of proteins with AlphaFold-derived representations"],"prefix":"10.1093","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1629-7756","authenticated-orcid":false,"given":"Maria","family":"Sindeeva","sequence":"first","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7280-1531","authenticated-orcid":false,"given":"Alexander","family":"Telepov","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0333-8117","authenticated-orcid":false,"given":"Nikita","family":"Ivanisenko","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8754-8727","authenticated-orcid":false,"given":"Tatiana","family":"Shashkova","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0446-6751","authenticated-orcid":false,"given":"Kuzma","family":"Khrabrov","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0754-759X","authenticated-orcid":false,"given":"Artem","family":"Tsypin","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1482-9365","authenticated-orcid":false,"given":"Artur","family":"Kadurin","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4827-8891","authenticated-orcid":false,"given":"Olga","family":"Kardymon","sequence":"additional","affiliation":[{"name":"Bioinformatics Group, AIRI , Moscow 121170 ,","place":["Russia"]}]}],"member":"286","published-online":{"date-parts":[[2025,7,7]]},"reference":[{"key":"2025081904112919200_ref1","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1016\/j.biotechadv.2011.09.008","article-title":"The imminent role of protein engineering in synthetic biology","volume":"30","author":"Foo","year":"2012","journal-title":"Biotechnol Adv"},{"key":"2025081904112919200_ref2","first-page":"14420","article-title":"Innovation by evolution: bringing new chemistry to life (nobel lecture)","volume-title":"Angew Chem Int Ed Engl","author":"Arnold","year":"2019"},{"key":"2025081904112919200_ref3","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1038\/nature19946","article-title":"The coming of age of de novo protein design","volume":"537","author":"Huang","year":"2016","journal-title":"Nature"},{"key":"2025081904112919200_ref4","doi-asserted-by":"crossref","first-page":"596","DOI":"10.1016\/j.sbi.2009.08.003","article-title":"Stability effects of mutations and protein evolvability","volume":"19","author":"Tokuriki","year":"2009","journal-title":"Curr Opin Struct Biol"},{"key":"2025081904112919200_ref5","doi-asserted-by":"crossref","first-page":"128","DOI":"10.1038\/nbt.3769","article-title":"Mutation effects predicted from sequence co-variation","volume":"35","author":"Hopf","year":"2017","journal-title":"Nat Biotechnol"},{"key":"2025081904112919200_ref6","doi-asserted-by":"crossref","first-page":"801","DOI":"10.1038\/nmeth.3027","article-title":"Deep mutational scanning: a new style of protein science","volume":"11","author":"Fowler","year":"2014","journal-title":"Nat Methods"},{"key":"2025081904112919200_ref7","doi-asserted-by":"crossref","first-page":"434","DOI":"10.1038\/s41586-023-06328-6","article-title":"Mega-scale experimental analysis of protein folding stability in biology and design","volume":"620","author":"Tsuboyama","year":"2023","journal-title":"Nature"},{"key":"2025081904112919200_ref8","doi-asserted-by":"crossref","DOI":"10.1101\/2024.08.12.606135","article-title":"The protein engineering tournament: an open science benchmark for protein modeling and design","volume-title":"ICLR 2024 Workshop on Generative and Experimental Perspectives for Biomolecular Design","author":"Armer","year":"2024"},{"key":"2025081904112919200_ref9","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein structure prediction with AlphaFold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"2025081904112919200_ref10","doi-asserted-by":"crossref","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate structure prediction of biomolecular interactions with AlphaFold 3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"key":"2025081904112919200_ref11","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1038\/s41592-022-01488-1","article-title":"ColabFold: making protein folding accessible to all","volume":"19","author":"Mirdita","year":"2022","journal-title":"Nat Methods"},{"article-title":"HelixFold: an efficient implementation of AlphaFold2 using paddlepaddle","year":"2022","author":"Wang","key":"2025081904112919200_ref12"},{"key":"2025081904112919200_ref13","doi-asserted-by":"crossref","DOI":"10.1038\/s41592-024-02272-z","article-title":"OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization","volume":"21","author":"Ahdritz","year":"2024","journal-title":"Nat Methods"},{"key":"2025081904112919200_ref14","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1126\/science.add2187","article-title":"Robust deep learning\u2013based protein sequence design using ProteinMPNN","volume":"378","author":"Dauparas","year":"2022","journal-title":"Science"},{"key":"2025081904112919200_ref15","doi-asserted-by":"crossref","first-page":"W382","DOI":"10.1093\/nar\/gki387","article-title":"The FoldX web server: an online force field","volume":"3","author":"Schymkowitz","year":"2005","journal-title":"Nucleic Acids Res"},{"key":"2025081904112919200_ref16","doi-asserted-by":"crossref","first-page":"3031","DOI":"10.1021\/acs.jctc.7b00125","article-title":"The Rosetta all-atom energy function for macromolecular modeling and design","volume":"13","author":"Alford","year":"2017","journal-title":"J Chem Theory Comput"},{"key":"2025081904112919200_ref17","article-title":"Embeddings from protein language models predict conservation and variant effects","volume":"141","author":"Marquet","year":"2021","journal-title":"Hum Genet"},{"key":"2025081904112919200_ref18","doi-asserted-by":"crossref","first-page":"btad671","DOI":"10.1093\/bioinformatics\/btad671","article-title":"PROSTATA: a framework for protein stability assessment using transformers","volume":"39","author":"Umerenkov","year":"2023","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref19","doi-asserted-by":"crossref","first-page":"e4780","DOI":"10.1002\/pro.4780","article-title":"Zero-shot mutation effect prediction on protein stability and function using RoseTTAFold","volume":"32","author":"Mansoor","year":"2023","journal-title":"Protein Sci"},{"key":"2025081904112919200_ref20","doi-asserted-by":"crossref","first-page":"2604","DOI":"10.1093\/molbev\/msz179","article-title":"GEMME: a simple and fast global epistatic model predicting mutational effects","volume":"36","author":"Laine","year":"2019","journal-title":"Mol Biol Evol"},{"key":"2025081904112919200_ref21","doi-asserted-by":"crossref","first-page":"e82593","DOI":"10.7554\/eLife.82593","article-title":"Rapid protein stability prediction using deep learning representations","volume":"12","author":"Blaabjerg","year":"2023","journal-title":"eLife"},{"key":"2025081904112919200_ref22","article-title":"Predicting a protein\u2019s stability under a million mutations","volume-title":"Advances in Neural Information Processing Systems","author":"Ouyang-Zhang","year":"2024"},{"key":"2025081904112919200_ref23","doi-asserted-by":"crossref","first-page":"6170","DOI":"10.1038\/s41467-024-49780-2","article-title":"Stability oracle: a structure-based graph-transformer framework for identifying stabilizing mutations","volume":"15","author":"Diaz","year":"2024","journal-title":"Nat Commun"},{"key":"2025081904112919200_ref24","doi-asserted-by":"crossref","first-page":"W338","DOI":"10.1093\/nar\/gkz383","article-title":"mCSM-PPI2: predicting the effects of mutations on protein\u2013protein interactions","volume":"47","author":"Rodrigues","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2025081904112919200_ref25","doi-asserted-by":"crossref","first-page":"e1009284","DOI":"10.1371\/journal.pcbi.1009284","article-title":"Deep geometric representations for modeling effects of mutations on protein\u2013protein binding affinity","volume":"17","author":"Liu","year":"2021","journal-title":"PLoS Comput Biol"},{"key":"2025081904112919200_ref26","doi-asserted-by":"crossref","first-page":"bbac024","DOI":"10.1093\/bib\/bbac024","article-title":"Persistent spectral based ensemble learning (PerSpect-EL) for protein\u2013protein binding affinity prediction","volume":"23","author":"Wee","year":"2022","journal-title":"Brief Bioinform"},{"key":"2025081904112919200_ref27","doi-asserted-by":"crossref","first-page":"107952","DOI":"10.1016\/j.compbiolchem.2023.107952","article-title":"ProS-GNN: predicting effects of mutations on protein stability using graph neural networks","volume":"107","author":"Wang","year":"2023","journal-title":"Comput Biol Chem"},{"key":"2025081904112919200_ref28","doi-asserted-by":"crossref","first-page":"bbad310","DOI":"10.1093\/bib\/bbad310","article-title":"MpbPPI: a multi-task pre-training-based equivariant approach for the prediction of the effect of amino acid mutations on protein\u2013protein interactions","volume":"24","author":"Yue","year":"2023","journal-title":"Brief Bioinform"},{"key":"2025081904112919200_ref29","first-page":"2024","article-title":"MuLAN: mutation-driven light attention networks for investigating protein\u2013protein interactions from sequences","author":"Lombardi","year":"2024"},{"key":"2025081904112919200_ref30","doi-asserted-by":"crossref","first-page":"116","DOI":"10.1038\/s42256-020-0149-6","article-title":"A topology-based network tree for the prediction of protein\u2013protein binding affinity changes following mutation","volume":"2","author":"Wang","year":"2020","journal-title":"Nat Mach Intell"},{"key":"2025081904112919200_ref31","doi-asserted-by":"crossref","first-page":"7400","DOI":"10.1038\/s41467-024-51776-x","article-title":"An end-to-end framework for the prediction of protein structure and fitness from single sequence","volume":"15","author":"Chen","year":"2024","journal-title":"Nat Commun"},{"key":"2025081904112919200_ref32","doi-asserted-by":"crossref","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"key":"2025081904112919200_ref33","first-page":"31","article-title":"PSnpBind-ML: predicting the effect of binding site mutations on protein-ligand binding affinity","volume":"15","author":"Ammar","year":"2023","journal-title":"J Chem"},{"key":"2025081904112919200_ref34","doi-asserted-by":"crossref","first-page":"btae621","DOI":"10.1093\/bioinformatics\/btae621","article-title":"Expert-guided protein language models enable accurate and blazingly fast fitness prediction","volume":"40","author":"Marquet","year":"2024","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref35","doi-asserted-by":"crossref","first-page":"238101","DOI":"10.1103\/PhysRevLett.129.238101","article-title":"State-of-the-art estimation of protein model accuracy using AlphaFold","volume":"129","author":"Roney","year":"2022","journal-title":"Phys Rev Lett"},{"key":"2025081904112919200_ref36","doi-asserted-by":"crossref","first-page":"e0282689","DOI":"10.1371\/journal.pone.0282689","article-title":"Using AlphaFold to predict the impact of single mutations on protein stability and function","volume":"18","author":"Pak","year":"2023","journal-title":"PloS One"},{"key":"2025081904112919200_ref37","doi-asserted-by":"crossref","first-page":"1508","DOI":"10.3390\/life14111508","article-title":"Design of ctenophore Ca+-regulated photoprotein berovin capable of being converted into active protein under physiological conditions: computational and experimental approaches","volume":"14","author":"Burakova","year":"2024","journal-title":"Life"},{"key":"2025081904112919200_ref38","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1038\/s41592-020-0848-2","article-title":"Macromolecular modeling and design in Rosetta: recent methods and frameworks","volume":"17","author":"Leman","year":"2020","journal-title":"Nat Methods"},{"key":"2025081904112919200_ref39","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1093\/nar\/28.1.235","article-title":"The Protein Data Bank","volume":"28","author":"Berman","year":"2000","journal-title":"Nucleic Acids Res"},{"key":"2025081904112919200_ref40","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4757-2440-0","volume-title":"The Nature of Statistical Learning Theory","author":"Vapnik","year":"1995"},{"key":"2025081904112919200_ref41","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach Learn"},{"key":"2025081904112919200_ref42","doi-asserted-by":"crossref","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy function approximation: a gradient boosting machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann Stat"},{"key":"2025081904112919200_ref43","first-page":"2825","article-title":"scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"2025081904112919200_ref44","doi-asserted-by":"crossref","first-page":"6201","DOI":"10.1021\/acs.jctc.6b00819","article-title":"Simultaneous optimization of biomolecular energy functions on features from small molecules and macromolecules","volume":"12","author":"Park","year":"2016","journal-title":"J Chem Theory Comput"},{"article-title":"Catboost: unbiased boosting with categorical features","volume-title":"Advances in Neural Information Processing Systems","author":"Prokhorenkova","key":"2025081904112919200_ref45"},{"key":"2025081904112919200_ref46","doi-asserted-by":"crossref","first-page":"584","DOI":"10.1006\/jmbi.1993.1413","article-title":"Prediction of protein secondary structure at better than 70% accuracy","volume":"232","author":"Rost","year":"1993","journal-title":"J Mol Biol"},{"key":"2025081904112919200_ref47","first-page":"149","article-title":"Using the Fisher kernel method to detect remote protein homologies","volume":"99","author":"Jaakkola","year":"1999","journal-title":"Proc Int Conf Intell Syst Mol Biol"},{"key":"2025081904112919200_ref48","doi-asserted-by":"crossref","DOI":"10.1101\/2022.12.31.522396","article-title":"New mega dataset combined with deep neural network makes a progress in predicting impact of mutation on protein stability","author":"Pak","year":"2023"},{"key":"2025081904112919200_ref49","doi-asserted-by":"crossref","first-page":"bbab555","DOI":"10.1093\/bib\/bbab555","article-title":"Predicting protein stability changes upon single-point mutation: a thorough comparison of the available tools on a new dataset","volume":"23","author":"Pancotti","year":"2022","journal-title":"Brief Bioinform"},{"key":"2025081904112919200_ref50","doi-asserted-by":"crossref","first-page":"3659","DOI":"10.1093\/bioinformatics\/bty348","article-title":"Quantification of biases in predictions of protein stability changes upon mutations","volume":"34","author":"Pucci","year":"2018","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref51","doi-asserted-by":"crossref","first-page":"01","DOI":"10.1186\/1471-2105-10-421","article-title":"Blast plus: architecture and applications","volume":"10","author":"Camacho","year":"2009","journal-title":"BMC Bioinformatics"},{"key":"2025081904112919200_ref52","doi-asserted-by":"crossref","first-page":"e46084","DOI":"10.1371\/journal.pone.0046084","article-title":"Assessing predictors of changes in protein stability upon mutation using self-consistency","volume":"7","author":"Thiltgen","year":"2012","journal-title":"PloS One"},{"key":"2025081904112919200_ref53","doi-asserted-by":"crossref","first-page":"2816","DOI":"10.1093\/bioinformatics\/btv291","article-title":"INPS: predicting the impact of non-synonymous variations on protein stability from sequence","volume":"31","author":"Fariselli","year":"2015","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref54","doi-asserted-by":"crossref","first-page":"3653","DOI":"10.1093\/bioinformatics\/bty340","article-title":"Self-consistency test reveals systematic bias in programs for prediction change of stability upon mutation","volume":"34","author":"Usmanova","year":"2018","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref55","doi-asserted-by":"crossref","first-page":"393","DOI":"10.1002\/pro.2829","article-title":"AB-bind: antibody binding mutational database for computational affinity predictions","volume":"25","author":"Sirin","year":"2015","journal-title":"Protein Sci"},{"key":"2025081904112919200_ref56","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1093\/bioinformatics\/bty635","article-title":"SKEMPI 2.0: an updated benchmark of changes in protein\u2013protein binding energy, kinetics and thermodynamics upon mutation","volume":"35","author":"Jankauskait\u0117","year":"2018","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref57","doi-asserted-by":"crossref","first-page":"i544","DOI":"10.1093\/bioinformatics\/btad231","article-title":"Deep local analysis deconstructs protein\u2013protein interfaces and accurately estimates binding affinity changes upon mutation","volume":"39","author":"Behbahani","year":"2023","journal-title":"Bioinformatics"},{"key":"2025081904112919200_ref58","doi-asserted-by":"crossref","first-page":"bbad491","DOI":"10.1093\/bib\/bbad491","article-title":"Quantification of biases in predictions of protein\u2013protein binding affinity changes upon mutations","volume":"25","author":"Tsishyn","year":"2024","journal-title":"Brief Bioinform"},{"key":"2025081904112919200_ref59","doi-asserted-by":"crossref","first-page":"D439","DOI":"10.1093\/nar\/gkab1061","article-title":"AlphaFold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models","volume":"50","author":"Varadi","year":"2022","journal-title":"Nucleic Acids Res"},{"key":"2025081904112919200_ref60","doi-asserted-by":"crossref","first-page":"D420","DOI":"10.1093\/nar\/gkaa1035","article-title":"ProThermDB: thermodynamic database for proteins and mutants revisited after 15 years","volume":"49","author":"Nikam","year":"2020","journal-title":"Nucleic Acids Res"},{"key":"2025081904112919200_ref61","doi-asserted-by":"crossref","first-page":"1017","DOI":"10.1093\/protein\/8.10.1017","article-title":"Enhancement of protein stability by the combination of point mutations in t4 lysozyme is additive","volume":"8","author":"Zhang","year":"1995","journal-title":"Protein Eng"},{"key":"2025081904112919200_ref62","article-title":"ProteinGym: large-scale benchmarks for protein fitness prediction and design","volume-title":"Advances in Neural Information Processing Systems","author":"Notin","year":"2024"},{"key":"2025081904112919200_ref63","article-title":"Attention is all you need","volume-title":"Advances in neural information processing systems","author":"Vaswani","year":"2017"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/4\/bbaf324\/63688281\/bbaf324.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/4\/bbaf324\/63688281\/bbaf324.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,19]],"date-time":"2025-08-19T08:11:40Z","timestamp":1755591100000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf324\/8190210"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":63,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,7,2]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf324","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"type":"print","value":"1467-5463"},{"type":"electronic","value":"1477-4054"}],"subject":[],"published-other":{"date-parts":[[2025,7]]},"published":{"date-parts":[[2025,7]]},"article-number":"bbaf324"}}