{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T08:08:29Z","timestamp":1773389309614,"version":"3.50.1"},"reference-count":34,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T00:00:00Z","timestamp":1773187200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"the Key Science and Technology Development Program of Henan Province","award":["252102210137"],"award-info":[{"award-number":["252102210137"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Accurate prediction of submitochondrial localization is fundamental to understanding mitochondrial biogenesis and cellular metabolic pathways. While deep representations from pre-trained protein language models (pLMs) have significantly advanced the field, traditional global average pooling methods often fail to capture critical, localized N-terminal targeting signals, particularly in long sequences where these motifs are mathematically diluted. To resolve this \u201csignal dilution\u201d bottleneck, we developed a multi-scale architecture that explicitly integrates high-resolution N-terminal features with global evolutionary context derived from ESM-2 embeddings. The proposed framework utilizes an orthogonal mixing strategy consisting of Token-mixing and Channel-mixing. Token-mixing is specifically designed to detect spatial rhythmic patterns across residue positions, while Channel-mixing refines the biochemical signatures within the latent feature space. Extensive benchmarking across diverse datasets demonstrates that our approach effectively maintains signal integrity. Compared to existing state-of-the-art methods, the model achieves a superior overall Generalized Correlation Coefficient (GCC) of 0.7443 on the SM424-18 dataset and 0.7878 on the SubMitoPred dataset, outperforming the latest benchmarks by 9.4% and 16.1%, respectively. Furthermore, on the independent M983 test set, our method maintained a high GCC of 0.6945, demonstrating a 9.9% improvement relative to the state-of-the-art methods. This robust and efficient framework provides a high-precision tool for large-scale mitochondrial proteomics.<\/jats:p>","DOI":"10.3390\/a19030212","type":"journal-article","created":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T14:43:18Z","timestamp":1773240198000},"page":"212","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Multi-Scale Feature Mixing of Language Model Embeddings for Enhanced Prediction of Submitochondrial Protein Localization"],"prefix":"10.3390","volume":"19","author":[{"given":"Rong","family":"Wang","sequence":"first","affiliation":[{"name":"School of Electronics and Information, Zhengzhou University of Light Industry, Zhengzhou 450002, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Menghua","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Zhengzhou University of Light Industry, Zhengzhou 450002, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yibo","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computational Economics, Henan University of Economics and Law, Zhengzhou 450046, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lixiang","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Zhengzhou University of Light Industry, Zhengzhou 450002, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3113-5149","authenticated-orcid":false,"given":"Xiao","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence, Zhengzhou University of Light Industry, Zhengzhou 450002, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,11]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"685","DOI":"10.1146\/annurev-biochem-060815-014352","article-title":"Mitochondrial machineries for protein import and assembly","volume":"86","author":"Wiedemann","year":"2017","journal-title":"Annu. Rev. Biochem."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"389","DOI":"10.1038\/nri2975","article-title":"Mitochondria in innate immune responses","volume":"11","author":"West","year":"2011","journal-title":"Nat. Rev. Immunol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"628","DOI":"10.1016\/j.cell.2009.08.005","article-title":"Importing mitochondrial proteins: Machineries and mechanisms","volume":"138","author":"Chacinska","year":"2009","journal-title":"Cell"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1038\/s41580-018-0092-0","article-title":"Mitochondrial proteins: From biogenesis to functional networks","volume":"20","author":"Pfanner","year":"2019","journal-title":"Nat. Rev. Mol. Cell Biol."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1038\/nrm.2017.66","article-title":"Mitochondrial diseases: The contribution of organelle stress responses to pathology","volume":"19","author":"Suomalainen","year":"2018","journal-title":"Nat. Rev. Mol. Cell Biol."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Huang, W.-L., Tung, C.-W., Ho, S.-W., Hwang, S.-F., and Ho, S.-Y. (2008). ProLoc-GO: Utilizing informative Gene Ontology terms for sequence-based prediction of protein subcellular localization. BMC Bioinform., 9.","DOI":"10.1186\/1471-2105-9-80"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1016\/j.biosystems.2009.06.007","article-title":"Predicting protein subnuclear localization using GO-amino-acid composition features","volume":"98","author":"Huang","year":"2009","journal-title":"Biosystems"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Kumar, R., Jain, S., Kumari, B., and Kumar, M. (2014). Protein sub-nuclear localization prediction using svm and pfam domain information. PLoS ONE, 9.","DOI":"10.1371\/journal.pone.0098345"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"30343","DOI":"10.3390\/ijms161226237","article-title":"Protein Sub-Nuclear Localization Based on Effective Fusion Representations and Dimension Reduction Algorithm LDA","volume":"16","author":"Wang","year":"2015","journal-title":"Int. J. Mol. Sci."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Wang, S., and Yue, Y. (2018). Protein subnuclear localization based on a new effective representation and intelligent kernel linear discriminant analysis by dichotomous greedy genetic algorithm. PLoS ONE, 13.","DOI":"10.1371\/journal.pone.0195636"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Littmann, M., Goldberg, T., Seitz, S., Bod\u00e9n, M., and Rost, B. (2019). Detailed prediction of protein sub-nuclear localization. BMC Bioinform., 20.","DOI":"10.1186\/s12859-019-2790-9"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2639","DOI":"10.1093\/bioinformatics\/btv212","article-title":"MultiP-SChlo: Multi-label protein subchloroplast localization prediction with Chou\u2019s pseudo amino acid composition and a novel multi-label classifier","volume":"31","author":"Wang","year":"2015","journal-title":"Bioinformatics"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"4755","DOI":"10.1021\/acs.jproteome.6b00686","article-title":"Ensemble Linear Neighborhood Propagation for Predicting Subchloroplast Localization of Multi-Location Proteins","volume":"15","author":"Wan","year":"2016","journal-title":"J. Proteome Res."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"212","DOI":"10.1109\/TCBB.2016.2527657","article-title":"Transductive Learning for Multi-Label Protein Subchloroplast Localization Prediction","volume":"14","author":"Wan","year":"2017","journal-title":"IEEE\/ACM Trans. Comput. Biol. Bioinform."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1449","DOI":"10.1109\/TCBB.2020.3037465","article-title":"An Effective Multi-Label Protein Sub-Chloroplast Localization Prediction by Skipped-Grams of Evolutionary Profiles Using Deep Neural Network","volume":"19","author":"Bankapur","year":"2022","journal-title":"IEEE\/ACM Trans. Comput. Biol. Bioinform."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"bbad083","DOI":"10.1093\/bib\/bbad083","article-title":"DaDL-SChlo: Protein subchloroplast localization prediction based on generative adversarial networks and pre-trained protein language model","volume":"24","author":"Wang","year":"2023","journal-title":"Brief. Bioinform."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Du, P., and Li, Y. (2006). Prediction of protein submitochondria locations by hybridizing pseudo-amino acid composition with various physicochemical features of segmented sequence. BMC Bioinform., 7.","DOI":"10.1186\/1471-2105-7-518"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1093\/bioinformatics\/btz512","article-title":"DeepMito: Accurate prediction of protein sub-mitochondrial localization using convolutional neural networks","volume":"36","author":"Savojardo","year":"2020","journal-title":"Bioinformatics"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"bbab288","DOI":"10.1093\/bib\/bbab288","article-title":"iDeepSubMito: Identification of protein submitochondrial localization with deep learning","volume":"22","author":"Hou","year":"2021","journal-title":"Brief. Bioinform."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1074","DOI":"10.1093\/bioinformatics\/btz734","article-title":"SubMito-XGBoost: Predicting protein submitochondrial localization by fusing multiple feature information and eXtreme gradient boosting","volume":"36","author":"Yu","year":"2020","journal-title":"Bioinformatics"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1016\/j.jtbi.2011.10.015","article-title":"Multi-kernel transfer learning based on Chou\u2019s PseAAC formulation for protein submitochondria localization","volume":"21","author":"Mei","year":"2012","journal-title":"J. Theor. Biol."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"e23270","DOI":"10.1096\/fj.202301019RR","article-title":"Subcellular localization and relevant mechanisms of human cancer-related micropeptides","volume":"37","author":"Deng","year":"2023","journal-title":"FASEB J."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"953","DOI":"10.1038\/nprot.2007.131","article-title":"Locating proteins in the cell using TargetP, SignalP and related tools","volume":"2","author":"Emanuelsson","year":"2007","journal-title":"Nat. Protoc."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"e201900429","DOI":"10.26508\/lsa.201900429","article-title":"Detecting sequence signals in targeting peptides using deep learning","volume":"2","author":"Armenteros","year":"2019","journal-title":"Life Sci. Alliance"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"263829","DOI":"10.1155\/2013\/263829","article-title":"SubMito-PSPCP: Predicting Protein Submitochondrial Locations by Hybridizing Positional Specific Physicochemical Properties with Pseudoamino Acid Compositions","volume":"2013","author":"Du","year":"2013","journal-title":"BioMed Res. Int."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"W459","DOI":"10.1093\/nar\/gky320","article-title":"BUSCA: An integrative web server to predict subcellular localization of proteins","volume":"46","author":"Savojardo","year":"2018","journal-title":"Nucleic Acids Res."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Wang, X., Jin, Y., and Zhang, Q. (2020). Deeppred-submito: A novel submitochondrial localization predictor based on multi-channel convolutional neural network and dataset balancing treatment. Int. J. Mol. Sci., 21.","DOI":"10.3390\/ijms21165710"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"4825","DOI":"10.1016\/j.csbj.2021.08.027","article-title":"MULocDeep: A deep-learning framework for protein subcellular and suborganellar localization prediction with residue-level interpretation","volume":"19","author":"Jiang","year":"2021","journal-title":"Comput. Struct. Biotechnol. J."},{"key":"ref_29","first-page":"24261","article-title":"MLP-Mixer: An all-mlp architecture for vision","volume":"34","author":"Tolstikhin","year":"2021","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1658","DOI":"10.1093\/bioinformatics\/btl158","article-title":"Cd-hit: A fast program for clustering and comparing large sets of protein or nucleotide sequences","volume":"22","author":"Li","year":"2006","journal-title":"Bioinformatics"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/j.mito.2017.10.004","article-title":"Proteome-wide prediction and annotation of mitochondrial and sub-mitochondrial proteins by incorporating domain information","volume":"42","author":"Kumar","year":"2018","journal-title":"Mitochondrion"},{"key":"ref_32","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":"ref_33","doi-asserted-by":"crossref","first-page":"412","DOI":"10.1093\/bioinformatics\/16.5.412","article-title":"Assessing the accuracy of prediction algorithms for classification: An overview","volume":"16","author":"Baldi","year":"2000","journal-title":"Bioinformatics"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Akiba, T., Sano, S., and Yanase, T. (2019, January 4\u20138). Optuna: A Next-generation Hyperparameter Optimization Framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, AK, USA.","DOI":"10.1145\/3292500.3330701"}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/3\/212\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T05:17:07Z","timestamp":1773379027000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/3\/212"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,11]]},"references-count":34,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["a19030212"],"URL":"https:\/\/doi.org\/10.3390\/a19030212","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,11]]}}}