{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T23:54:08Z","timestamp":1790898848279,"version":"4.1.0"},"reference-count":44,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T00:00:00Z","timestamp":1788307200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0"},{"start":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T00:00:00Z","timestamp":1790899200000},"content-version":"vor","delay-in-days":30,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Nat Commun"],"DOI":"10.1038\/s41467-026-77333-2","type":"journal-article","created":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T13:14:57Z","timestamp":1788354897000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Unsupervised discovery of functional sequence patterns from protein language model with MotifAE"],"prefix":"10.1038","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4806-7637","authenticated-orcid":false,"given":"Chao","family":"Hou","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-2021-4468","authenticated-orcid":false,"given":"Di","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1299-5979","authenticated-orcid":false,"given":"Yufeng","family":"Shen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,9,2]]},"reference":[{"key":"77333_CR1","doi-asserted-by":"publisher","first-page":"D442","DOI":"10.1093\/nar\/gkad1058","volume":"52","author":"M Kumar","year":"2024","unstructured":"Kumar, M. et al. ELM-the Eukaryotic Linear Motif resource-2024 update. Nucleic Acids Res. 52, D442\u2013D455 (2024).","journal-title":"Nucleic Acids Res."},{"key":"77333_CR2","doi-asserted-by":"publisher","first-page":"1292","DOI":"10.1016\/j.molcel.2021.01.020","volume":"81","author":"C Tokheim","year":"2021","unstructured":"Tokheim, C. et al. Systematic characterization of mutations altering protein degradation in human cancers. Mol. Cell 81, 1292\u20131308.e11 (2021).","journal-title":"Mol. Cell"},{"key":"77333_CR3","doi-asserted-by":"publisher","first-page":"102641","DOI":"10.1016\/j.sbi.2023.102641","volume":"81","author":"C Savojardo","year":"2023","unstructured":"Savojardo, C., Martelli, P. L. & Casadio, R. Finding functional motifs in protein sequences with deep learning and natural language models. Curr. Opin. Struct. Biol. 81, 102641 (2023).","journal-title":"Curr. Opin. Struct. Biol."},{"key":"77333_CR4","doi-asserted-by":"publisher","DOI":"10.1186\/s12915-022-01364-6","volume":"20","author":"C Hou","year":"2022","unstructured":"Hou, C., Li, Y., Wang, M., Wu, H. & Li, T. Systematic prediction of degrons and E3 ubiquitin ligase binding via deep learning. BMC Biol. 20, 162 (2022).","journal-title":"BMC Biol."},{"key":"77333_CR5","doi-asserted-by":"publisher","first-page":"235","DOI":"10.1093\/nar\/28.1.235","volume":"28","author":"HM Berman","year":"2000","unstructured":"Berman, H. M. et al. The protein data bank. Nucleic Acids Res. 28, 235\u2013242 (2000).","journal-title":"Nucleic Acids Res."},{"key":"77333_CR6","doi-asserted-by":"crossref","unstructured":"Lau, A. M. et al. Exploring structural diversity across the protein universe with the encyclopedia of domains. Science 386, eadq4946 (2024).","DOI":"10.1126\/science.adq4946"},{"key":"77333_CR7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmb.2024.168551","volume":"436","author":"VP Waman","year":"2024","unstructured":"Waman, V. P. et al. CATH 2024: CATH-alphaflow doubles the number of structures in CATH and Reveals Nearly 200 New Folds. J. Mol. Biol. 436, 168551 (2024).","journal-title":"J. Mol. Biol."},{"key":"77333_CR8","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1126\/science.ade2574","volume":"379","author":"Z Lin","year":"2023","unstructured":"Lin, Z. et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science 379, 1123\u20131130 (2023).","journal-title":"Science"},{"key":"77333_CR9","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","volume":"596","author":"J Jumper","year":"2021","unstructured":"Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583\u2013589\u00a0(2021).","journal-title":"Nature"},{"key":"77333_CR10","doi-asserted-by":"publisher","first-page":"220","DOI":"10.1038\/s42256-024-00795-w","volume":"6","author":"M Kulmanov","year":"2024","unstructured":"Kulmanov, M. et al. Protein function prediction as approximate semantic entailment. Nat. Mach. Intell. 6, 220\u2013228 (2024).","journal-title":"Nat. Mach. Intell."},{"key":"77333_CR11","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2406285121","volume":"121","author":"Z Zhang","year":"2024","unstructured":"Zhang, Z. et al. Protein language models learn evolutionary statistics of interacting sequence motifs. Proc. Natl. Acad. Sci. USA 121, e2406285121 (2024).","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"77333_CR12","doi-asserted-by":"crossref","unstructured":"Vig, J. et al. BERTology meets biology: interpreting attention in protein language models. Preprint at https:\/\/arxiv.org\/abs\/2006.15222 (2020).","DOI":"10.1101\/2020.06.26.174417"},{"key":"77333_CR13","doi-asserted-by":"crossref","unstructured":"Rives, A. et al. Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. Proc. Natl. Acad. Sci. USA 118, e2016239118 (2021).","DOI":"10.1073\/pnas.2016239118"},{"key":"77333_CR14","unstructured":"Huben, R., Cunningham, H., Smith, L. R., Ewart, A. & Sharkey, L. Sparse autoencoders find highly interpretable features in language models. ICLR (2024)."},{"key":"77333_CR15","doi-asserted-by":"crossref","unstructured":"Yun, Z., Chen, Y., Olshausen, B. A. & LeCun, Y. Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors. DeeLIO 1\u201310 (ACL, 2021).","DOI":"10.18653\/v1\/2021.deelio-1.1"},{"key":"77333_CR16","doi-asserted-by":"crossref","unstructured":"Adams, E., Bai, L., Lee, M., Yu, Y. & AlQuraishi, M. From mechanistic interpretability to mechanistic biology: training, evaluating, and interpreting sparse autoencoders on protein language models. ICML 460\u2013476 (PMLR, 2025).","DOI":"10.1101\/2025.02.06.636901"},{"key":"77333_CR17","doi-asserted-by":"publisher","first-page":"2107","DOI":"10.1038\/s41592-025-02836-7","volume":"22","author":"E Simon","year":"2025","unstructured":"Simon, E. & Zou, J. InterPLM: discovering interpretable features in protein language models via sparse autoencoders. Nat. Methods 22, 2107\u20132117 (2025).","journal-title":"Nat. Methods"},{"key":"77333_CR18","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2506316122","volume":"122","author":"O Gujral","year":"2025","unstructured":"Gujral, O., Bafna, M., Alm, E. & Berger, B. Sparse autoencoders uncover biologically interpretable features in protein language model representations. Proc. Natl. Acad. Sci. 122, e2506316122 (2025).","journal-title":"Proc. Natl. Acad. Sci."},{"key":"77333_CR19","doi-asserted-by":"publisher","first-page":"1512","DOI":"10.1038\/s41588-023-01465-0","volume":"55","author":"N Brandes","year":"2023","unstructured":"Brandes, N., Goldman, G., Wang, C. H., Ye, C. J. & Ntranos, V. Genome-wide prediction of disease variant effects with a deep protein language model. Nat. Genet 55, 1512\u20131522 (2023).","journal-title":"Nat. Genet"},{"key":"77333_CR20","first-page":"64331","volume":"36","author":"P Notin","year":"2023","unstructured":"Notin, P. et al. ProteinGym: large-scale benchmarks for protein fitness prediction and design.\u00a0NeurIPS 36, 64331\u201364379 (2023).","journal-title":"NeurIPS"},{"key":"77333_CR21","doi-asserted-by":"crossref","unstructured":"Hou, C., Liu, D., Zafar, A. & Shen, Y. Understanding Language Model Scaling for Protein Fitness Prediction. Nat. Comput. Sci. 6, 778\u2013788 (2026).","DOI":"10.1038\/s43588-026-01010-z"},{"key":"77333_CR22","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1038\/s41586-023-06510-w","volume":"622","author":"I Barrio-Hernandez","year":"2023","unstructured":"Barrio-Hernandez, I. et al. Clustering predicted structures at the scale of the known protein universe. Nature 622, 637\u2013645 (2023).","journal-title":"Nature"},{"key":"77333_CR23","doi-asserted-by":"publisher","first-page":"D1313","DOI":"10.1093\/nar\/gkae1090","volume":"53","author":"MJ Landrum","year":"2025","unstructured":"Landrum, M. J. et al. ClinVar: updates to support classifications of both germline and somatic variants. Nucleic Acids Res. 53, D1313\u2013D1321 (2025).","journal-title":"Nucleic Acids Res."},{"key":"77333_CR24","doi-asserted-by":"publisher","first-page":"1385","DOI":"10.7150\/ijbs.5141","volume":"8","author":"M Okada","year":"2012","unstructured":"Okada, M. Regulation of the Src family kinases by Csk. Int J. Biol. Sci. 8, 1385\u20131397 (2012).","journal-title":"Int J. Biol. Sci."},{"key":"77333_CR25","doi-asserted-by":"publisher","first-page":"W458","DOI":"10.1093\/nar\/gkx248","volume":"45","author":"M Andreatta","year":"2017","unstructured":"Andreatta, M., Alvarez, B. & Nielsen, M. GibbsCluster: unsupervised clustering and alignment of peptide sequences. Nucleic Acids Res 45, W458\u2013W463 (2017).","journal-title":"Nucleic Acids Res"},{"key":"77333_CR26","doi-asserted-by":"publisher","first-page":"434","DOI":"10.1038\/s41586-023-06328-6","volume":"620","author":"K Tsuboyama","year":"2023","unstructured":"Tsuboyama, K. et al. Mega-scale experimental analysis of protein folding stability in biology and design. Nature 620, 434\u2013444 (2023).","journal-title":"Nature"},{"key":"77333_CR27","doi-asserted-by":"crossref","unstructured":"van Kempen, M. et al. Fast and accurate protein structure search with Foldseek. Nat. Biotechnol. 42, 243\u2013246 (2023).","DOI":"10.1038\/s41587-023-01773-0"},{"key":"77333_CR28","doi-asserted-by":"publisher","first-page":"W382","DOI":"10.1093\/nar\/gki387","volume":"33","author":"J Schymkowitz","year":"2005","unstructured":"Schymkowitz, J. et al. The FoldX web server: an online force field. Nucleic Acids Res. 33, W382\u2013W388 (2005).","journal-title":"Nucleic Acids Res."},{"key":"77333_CR29","doi-asserted-by":"crossref","unstructured":"Darmawan, J. T., Gal, Y. & Notin, P. Sampling protein language models for functional protein design. ICLR LMRL Workshop (2025).","DOI":"10.1101\/2025.09.14.676087"},{"key":"77333_CR30","unstructured":"Gao, L. et al. Scaling and evaluating sparse autoencoders. ICLR (2025)."},{"key":"77333_CR31","unstructured":"Bussmann, B., Leask, P. & Nanda, N. BatchTopK Sparse Autoencoders. NeurIPS SciForDL Workshop (2024)."},{"key":"77333_CR32","unstructured":"Paszke, A. et al. PyTorch: an imperative style, high-performance deep learning library. NeurIPS 32, 8024\u20138035 (2019)."},{"key":"77333_CR33","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-62318-4","volume":"16","author":"G Zhong","year":"2025","unstructured":"Zhong, G., Zhao, Y., Zhuang, D., Chung, W. K. & Shen, Y. PreMode predicts mode-of-action of missense variants by deep graph representation learning of protein sequence and structural context. Nat. Commun. 16, 7189 (2025).","journal-title":"Nat. Commun."},{"key":"77333_CR34","doi-asserted-by":"crossref","unstructured":"Cheng, J. et al. Accurate proteome-wide missense variant effect prediction with AlphaMissense. Science 381, eadg7492 (2023).","DOI":"10.1126\/science.adg7492"},{"key":"77333_CR35","doi-asserted-by":"publisher","first-page":"1528","DOI":"10.1016\/j.bpj.2015.08.015","volume":"109","author":"RT McGibbon","year":"2015","unstructured":"McGibbon, R. T. et al. MDTraj: a modern open library for the analysis of molecular dynamics trajectories. Biophys. J. 109, 1528\u20131532 (2015).","journal-title":"Biophys. J."},{"key":"77333_CR36","doi-asserted-by":"publisher","first-page":"W297","DOI":"10.1093\/nar\/gkab408","volume":"49","author":"G Erdos","year":"2021","unstructured":"Erdos, G., Pajkos, M. & Dosztanyi, Z. IUPred3: prediction of protein disorder enhanced with unambiguous experimental annotation and visualization of evolutionary conservation. Nucleic acids Res. 49, W297\u2013W303 (2021).","journal-title":"Nucleic acids Res."},{"key":"77333_CR37","doi-asserted-by":"crossref","unstructured":"The UniProt Consortium. UniProt: a worldwide hub of protein knowledge. Nucleic Acids Res. 47, D506\u2013D515 (2019).","DOI":"10.1093\/nar\/gky1049"},{"key":"77333_CR38","doi-asserted-by":"publisher","first-page":"2272","DOI":"10.1093\/bioinformatics\/btz921","volume":"36","author":"A Tareen","year":"2020","unstructured":"Tareen, A. & Kinney, J. B. Logomaker: beautiful sequence logos in Python. Bioinformatics 36, 2272\u20132274 (2020).","journal-title":"Bioinformatics"},{"key":"77333_CR39","unstructured":"Yin, P. et al. Understanding straight-through estimator in training activation quantized neural nets. ICLR (2019)."},{"key":"77333_CR40","unstructured":"Blondel, M., Teboul, O., Berthet, Q. & Djolonga, J. Fast differentiable sorting and ranking. ICML 950\u2013959 (PMLR, 2020)."},{"key":"77333_CR41","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2314853121","volume":"121","author":"H Dieckhaus","year":"2024","unstructured":"Dieckhaus, H., Brocidiacono, M., Randolph, N. Z. & Kuhlman, B. Transfer learning to leverage larger datasets for improved prediction of protein stability changes. Proc. Natl. Acad. Sci. USA 121, e2314853121 (2024).","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"77333_CR42","doi-asserted-by":"publisher","first-page":"1099","DOI":"10.1038\/s41587-022-01618-2","volume":"41","author":"A Madani","year":"2023","unstructured":"Madani, A. et al. Large language models generate functional protein sequences across diverse families. Nat. Biotechnol. 41, 1099\u20131106 (2023).","journal-title":"Nat. Biotechnol."},{"key":"77333_CR43","doi-asserted-by":"publisher","unstructured":"Tsuboyama, K. et al. Mega-scale experimental analysis of protein folding stability in biology and design. https:\/\/doi.org\/10.5281\/zenodo.7992926 (2023).","DOI":"10.5281\/zenodo.7992926"},{"key":"77333_CR44","doi-asserted-by":"publisher","unstructured":"Hou, C. Data and model weights for MotifAE, a sparse autoencoder for unsupervised discovery of functional motifs from protein language model. https:\/\/doi.org\/10.5281\/zenodo.20547228 (2026).","DOI":"10.5281\/zenodo.20547228"}],"container-title":["Nature Communications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-77333-2","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-77333-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-77333-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T23:01:06Z","timestamp":1790895666000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41467-026-77333-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,2]]},"references-count":44,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["77333"],"URL":"https:\/\/doi.org\/10.1038\/s41467-026-77333-2","relation":{},"ISSN":["2041-1723"],"issn-type":[{"value":"2041-1723","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,2]]},"assertion":[{"value":"26 February 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 August 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 September 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"10444"}}