{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,22]],"date-time":"2026-08-22T20:16:13Z","timestamp":1787429773681,"version":"3.56.0"},"reference-count":35,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,9,19]],"date-time":"2023-09-19T00:00:00Z","timestamp":1695081600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,9,19]],"date-time":"2023-09-19T00:00:00Z","timestamp":1695081600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["22274151"],"award-info":[{"award-number":["22274151"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["22174141"],"award-info":[{"award-number":["22174141"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["22074145"],"award-info":[{"award-number":["22074145"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["21934006"],"award-info":[{"award-number":["21934006"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"The Innovation Program of Science and Research from Dalian Institute of Chemical Physics, CAS, China","award":["DICP I202004"],"award-info":[{"award-number":["DICP I202004"]}]},{"name":"The AI S&T Program from Yulin Branch, Dalian National Laboratory For Clean Energy, CAS, China","award":["DNL-YL A202202"],"award-info":[{"award-number":["DNL-YL A202202"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Bioinformatics"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Background<\/jats:title>\n                <jats:p>Plant secondary metabolites are highly valued for their applications in pharmaceuticals, nutrition, flavors, and aesthetics. It is of great importance to elucidate plant secondary metabolic pathways due to their crucial roles in biological processes during plant growth and development. However, understanding plant biosynthesis and degradation pathways remains a challenge due to the lack of sufficient information in current databases. To address this issue, we proposed a transfer learning approach using a pre-trained hybrid deep learning architecture that combines Graph Transformer and convolutional neural network (GTC) to predict plant metabolic pathways.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>GTC provides comprehensive molecular representation by extracting both structural features from the molecular graph and textual information from the SMILES string. GTC is pre-trained on the KEGG datasets to acquire general features, followed by fine-tuning on plant-derived datasets. Four metrics were chosen for model performance evaluation. The results show that GTC outperforms six other models, including three previously reported machine learning models, on the KEGG dataset. GTC yields an accuracy of 96.75%, precision of 85.14%, recall of 83.03%, and F1_score of 84.06%. Furthermore, an ablation study confirms the indispensability of all the components of the hybrid GTC model. Transfer learning is then employed to leverage the shared knowledge acquired from the KEGG metabolic pathways. As a result, the transferred GTC exhibits outstanding accuracy in predicting plant secondary metabolic pathways with an average accuracy of 98.30% in fivefold cross-validation and 97.82% on the final test. In addition, GTC is employed to classify natural products. It achieves a perfect accuracy score of 100.00% for alkaloids, while the lowest accuracy score of 98.42% for shikimates and phenylpropanoids.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusions<\/jats:title>\n                <jats:p>The proposed GTC effectively captures molecular features, and achieves high performance in classifying KEGG metabolic pathways and predicting plant secondary metabolic pathways via transfer learning. Furthermore, GTC demonstrates its generalization ability by accurately classifying natural products. A user-friendly executable program has been developed, which only requires the input of the SMILES string of the query compound in a graphical interface.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1186\/s12859-023-05485-9","type":"journal-article","created":{"date-parts":[[2023,9,19]],"date-time":"2023-09-19T06:03:31Z","timestamp":1695103411000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":22,"title":["Prediction of plant secondary metabolic pathways using deep transfer learning"],"prefix":"10.1186","volume":"24","author":[{"given":"Han","family":"Bao","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinhui","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinjie","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chunxia","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5569-1740","authenticated-orcid":false,"given":"Xin","family":"Lu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4298-3554","authenticated-orcid":false,"given":"Guowang","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,9,19]]},"reference":[{"key":"5485_CR1","doi-asserted-by":"publisher","first-page":"977","DOI":"10.1016\/j.plaphy.2005.07.013","volume":"43","author":"SMA Zobayed","year":"2005","unstructured":"Zobayed SMA, Afreen F, Kozai T. Temperature stress can alter the photosynthetic efficiency and secondary metabolite concentrations in St. John\u2019s wort. Plant Physiol Biochem. 2005;43:977\u201384.","journal-title":"Plant Physiol Biochem"},{"key":"5485_CR2","doi-asserted-by":"publisher","first-page":"447","DOI":"10.3390\/plants12030447","volume":"12","author":"MM Qaderi","year":"2023","unstructured":"Qaderi MM, Martel AB, Strugnell CA. Environmental factors regulate plant secondary metabolites. Plants. 2023;12:447.","journal-title":"Plants"},{"key":"5485_CR3","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1016\/j.mycmed.2017.04.008","volume":"27","author":"F Lagrouh","year":"2017","unstructured":"Lagrouh F, Dakka N, Bakri Y. The antifungal activity of Moroccan plants and the mechanism of action of secondary metabolites from plants. J de Mycol M\u00e9d. 2017;27:303\u201311.","journal-title":"J de Mycol M\u00e9d"},{"key":"5485_CR4","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1007\/s13205-020-02541-6","volume":"10","author":"J Mishra","year":"2020","unstructured":"Mishra J, Srivastava R, Trivedi PK, Verma PC. Effect of virus infection on the secondary metabolite production and phytohormone biosynthesis in plants. 3 Biotech. 2020;10:547.","journal-title":"3 Biotech"},{"key":"5485_CR5","doi-asserted-by":"publisher","first-page":"12570","DOI":"10.3390\/ijms222212570","volume":"22","author":"S Fakhri","year":"2021","unstructured":"Fakhri S, Abdian S, Zarneshan SN, Akkol EK, Farzaei MH, Sobarzo-S\u00e1nchez E. Targeting mitochondria by plant secondary metabolites: a promising strategy in combating Parkinson\u2019s disease. Int J Mol Sci. 2021;22:12570.","journal-title":"Int J Mol Sci"},{"key":"5485_CR6","doi-asserted-by":"publisher","first-page":"1076","DOI":"10.3390\/metabo3041076","volume":"3","author":"R Breitling","year":"2013","unstructured":"Breitling R, Ceniceros A, Jankevics A, Takano E. Metabolomics for secondary metabolite research. Metabolites. 2013;3:1076\u201383.","journal-title":"Metabolites"},{"key":"5485_CR7","doi-asserted-by":"publisher","first-page":"423","DOI":"10.1016\/j.foodchem.2016.11.132","volume":"224","author":"J Rubert","year":"2017","unstructured":"Rubert J, Righetti L, Stranska-Zachariasova M, Dzuman Z, Chrpova J, Dall\u2019Asta C, et al. Untargeted metabolomics based on ultra-high-performance liquid chromatography\u2013high-resolution mass spectrometry merged with chemometrics: a new predictable tool for an early detection of mycotoxins. Food Chem. 2017;224:423\u201331.","journal-title":"Food Chem"},{"key":"5485_CR8","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1111\/tpj.14727","volume":"103","author":"T Shi","year":"2020","unstructured":"Shi T, Zhu A, Jia J, Hu X, Chen J, Liu W, et al. Metabolomics analysis and metabolite-agronomic trait associations using kernels of wheat (Triticum aestivum) recombinant inbred lines. Plant J. 2020;103:279\u201392.","journal-title":"Plant J"},{"key":"5485_CR9","doi-asserted-by":"publisher","DOI":"10.1016\/j.pbi.2021.102171","volume":"66","author":"P Wang","year":"2022","unstructured":"Wang P, Schumacher AM, Shiu S-H. Computational prediction of plant metabolic pathways. Curr Opin Plant Biol. 2022;66: 102171.","journal-title":"Curr Opin Plant Biol"},{"key":"5485_CR10","doi-asserted-by":"publisher","first-page":"530","DOI":"10.1093\/bib\/bbq083","volume":"12","author":"Y Sucaet","year":"2011","unstructured":"Sucaet Y, Deva T. Evolution and applications of plant pathway resources and databases. Brief Bioinform. 2011;12:530\u201344.","journal-title":"Brief Bioinform"},{"key":"5485_CR11","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1007\/s11030-008-9085-9","volume":"12","author":"Y-D Cai","year":"2008","unstructured":"Cai Y-D, Qian Z, Lu L, Feng K-Y, Meng X, Niu B, et al. Prediction of compounds\u2019 biological function (metabolic pathways) based on functional group composition. Mol Divers. 2008;12:131\u20137.","journal-title":"Mol Divers"},{"key":"5485_CR12","doi-asserted-by":"publisher","first-page":"969","DOI":"10.2174\/092986609788923374","volume":"16","author":"J Lu","year":"2009","unstructured":"Lu J, Niu B, Liu L, Lu W-C, Cai Y-D. Prediction of small molecules\u2019 metabolic pathways based on functional group composition. Protein Pept Lett. 2009;16:969\u201376.","journal-title":"Protein Pept Lett"},{"key":"5485_CR13","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0029491","volume":"6","author":"L-L Hu","year":"2011","unstructured":"Hu L-L, Chen C, Huang T, Cai Y-D, Chou K-C. Predicting biological functions of compounds based on chemical-chemical interactions. PLoS ONE. 2011;6: e29491.","journal-title":"PLoS ONE"},{"key":"5485_CR14","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0045944","volume":"7","author":"Y-F Gao","year":"2012","unstructured":"Gao Y-F, Chen L, Cai Y-D, Feng K-Y, Huang T, Jiang Y. Predicting metabolic pathways of small molecules and enzymes based on interaction information of chemicals and proteins. PLoS ONE. 2012;7: e45944.","journal-title":"PLoS ONE"},{"key":"5485_CR15","doi-asserted-by":"publisher","first-page":"709","DOI":"10.1021\/ci500517v","volume":"55","author":"MA Hamdalla","year":"2015","unstructured":"Hamdalla MA, Rajasekaran S, Grant DF, M\u0103ndoiu II. Metabolic pathway predictions for metabolomics: a molecular structure matching approach. J Chem Inf Model. 2015;55:709\u201318.","journal-title":"J Chem Inf Model"},{"key":"5485_CR16","volume":"2021","author":"Y Wang","year":"2021","unstructured":"Wang Y, Xu Y, Yang Z, Liu X, Dai Q. Using recursive feature selection with random forest to improve protein structural class prediction for low-similarity sequences. Comput Math Methods Med. 2021;2021: e5529389.","journal-title":"Comput Math Methods Med"},{"key":"5485_CR17","doi-asserted-by":"publisher","first-page":"136","DOI":"10.2174\/1386207319666151110122453","volume":"19","author":"L Chen","year":"2016","unstructured":"Chen L, Chu C, Feng K. Predicting the types of metabolic pathway of compounds using molecular fragments and sequential minimal optimization. Comb Chem High Throughput Screen. 2016;19:136\u201343.","journal-title":"Comb Chem High Throughput Screen"},{"key":"5485_CR18","doi-asserted-by":"publisher","first-page":"670","DOI":"10.2174\/1386207322666181206112641","volume":"21","author":"Z-H Guo","year":"2018","unstructured":"Guo Z-H, Chen L, Zhao X. A network integration method for deciphering the types of metabolic pathway of chemicals with heterogeneous information. Comb Chem High Throughput Screen. 2018;21:670\u201380.","journal-title":"Comb Chem High Throughput Screen"},{"key":"5485_CR19","doi-asserted-by":"publisher","first-page":"83","DOI":"10.2174\/1875036202013010083","volume":"13","author":"Y Jia","year":"2020","unstructured":"Jia Y, Chen L, Zhou J-P, Liu M. iMPT-FRAKEL: a simple multi-label web-server that only uses fingerprints to identify which metabolic pathway types compounds can participate in. Open Bioinform J. 2020;13:83.","journal-title":"Open Bioinform J"},{"key":"5485_CR20","doi-asserted-by":"publisher","DOI":"10.1155\/2021\/6683051","volume":"2021","author":"Y Zhu","year":"2021","unstructured":"Zhu Y, Hu B, Chen L, Dai Q. iMPTCE-Hnetwork: a multilabel classifier for identifying metabolic pathway types of chemicals and enzymes with a heterogeneous network. Comput Math Methods Med. 2021;2021: e6683051.","journal-title":"Comput Math Methods Med"},{"key":"5485_CR21","doi-asserted-by":"publisher","first-page":"2547","DOI":"10.1093\/bioinformatics\/btz954","volume":"36","author":"M Baranwal","year":"2020","unstructured":"Baranwal M, Magner A, Elvati P, Saldinger J, Violi A, Hero AO. A deep learning architecture for metabolic pathway prediction. Bioinformatics. 2020;36:2547\u201353.","journal-title":"Bioinformatics"},{"key":"5485_CR22","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1021\/ci00057a005","volume":"28","author":"D Weininger","year":"1988","unstructured":"Weininger D. SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules. J Chem Inf Comput Sci. 1988;28:31\u20136.","journal-title":"J Chem Inf Comput Sci"},{"key":"5485_CR23","doi-asserted-by":"crossref","unstructured":"Yang Z, Liu J, Wang Z, Wang Y, Feng J. Multi-class metabolic pathway prediction by graph attention-based deep learning method. In: 2020 IEEE International conference on bioinformatics and biomedicine (BIBM). 2020:126\u201331.","DOI":"10.1109\/BIBM49941.2020.9313298"},{"issue":"Supplement_1","key":"5485_CR24","doi-asserted-by":"publisher","first-page":"i325","DOI":"10.1093\/bioinformatics\/btac222","volume":"38","author":"B-X Du","year":"2022","unstructured":"Du B-X, Zhao P-C, Zhu B, Yiu S-M, Nyamabo AK, Yu H, et al. MLGL-MP: a multi-label graph learning framework enhanced by pathway interdependence for metabolic pathway prediction. Bioinformatics. 2022;38(Supplement_1):i325\u201332.","journal-title":"Bioinformatics"},{"key":"5485_CR25","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1093\/nar\/28.1.27","volume":"28","author":"M Kanehisa","year":"2000","unstructured":"Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28:27\u201330.","journal-title":"Nucleic Acids Res"},{"key":"5485_CR26","doi-asserted-by":"publisher","first-page":"130687","DOI":"10.1109\/ACCESS.2020.3009439","volume":"8","author":"Y Jia","year":"2020","unstructured":"Jia Y, Zhao R, Chen L. Similarity-based machine learning model for predicting the metabolic pathways of compounds. IEEE Access. 2020;8:130687\u201396.","journal-title":"IEEE Access"},{"key":"5485_CR27","doi-asserted-by":"crossref","unstructured":"Shi Y, Huang Z, Feng S, Zhong H, Wang W, Sun Y. Masked label prediction: unified message passing model for semi-supervised classification. 2021. Preprint arXiv:2009.03509.","DOI":"10.24963\/ijcai.2021\/214"},{"key":"5485_CR28","doi-asserted-by":"publisher","first-page":"1479","DOI":"10.1104\/pp.110.157396","volume":"153","author":"P Zhang","year":"2010","unstructured":"Zhang P, Dreher K, Karthikeyan A, Chi A, Pujar A, Caspi R, et al. Creation of a genome-wide metabolic pathway database for Populus trichocarpa using a new approach for reconstruction and curation of metabolic pathways for plants. Plant Physiol. 2010;153:1479\u201391.","journal-title":"Plant Physiol"},{"key":"5485_CR29","doi-asserted-by":"publisher","first-page":"2795","DOI":"10.1021\/acs.jnatprod.1c00399","volume":"84","author":"HW Kim","year":"2021","unstructured":"Kim HW, Wang M, Leber CA, Nothias L-F, Reher R, Kang KB, et al. NPClassifier: a deep neural network-based structural classification tool for natural products. J Nat Prod. 2021;84:2795\u2013807.","journal-title":"J Nat Prod"},{"key":"5485_CR30","doi-asserted-by":"publisher","first-page":"1140","DOI":"10.1093\/bioinformatics\/btaa921","volume":"37","author":"T Nguyen","year":"2021","unstructured":"Nguyen T, Le H, Quinn TP, Nguyen T, Le TD, Venkatesh S. GraphDTA: predicting drug\u2013target binding affinity with graph neural networks. Bioinformatics. 2021;37:1140\u20137.","journal-title":"Bioinformatics"},{"key":"5485_CR31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/1758-2946-5-26","volume":"5","author":"S Riniker","year":"2013","unstructured":"Riniker S, Landrum GA. Open-source platform to benchmark fingerprints for ligand-based virtual screening. J Cheminform. 2013;5:1\u201317.","journal-title":"J Cheminform"},{"key":"5485_CR32","volume-title":"Advances in neural information processing systems","author":"A Vaswani","year":"2017","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. In: Cortes C, Lawarence N, Lee D, Sugiyama M, Garnett R, editors., et al., Advances in neural information processing systems. New York: Curran Associates Inc.; 2017."},{"key":"5485_CR33","doi-asserted-by":"publisher","first-page":"4191","DOI":"10.1021\/acs.jcim.9b01212","volume":"60","author":"P Morris","year":"2020","unstructured":"Morris P, St. Clair R, Hahn WE, Barenholtz E. Predicting binding from screening assays with transformer network embeddings. J Chem Inf Model. 2020;60:4191\u20139.","journal-title":"J Chem Inf Model"},{"key":"5485_CR34","unstructured":"Kim D, Oh A. How to find your friendly neighborhood: graph attention design with self-supervision. 2023. Preprint arXiv:2204.04879."},{"key":"5485_CR35","doi-asserted-by":"publisher","first-page":"2","DOI":"10.1186\/s13321-020-00478-9","volume":"13","author":"M Sorokina","year":"2021","unstructured":"Sorokina M, Merseburger P, Rajan K, Yirik MA, Steinbeck C. COCONUT online: collection of open natural products database. J Cheminform. 2021;13:2.","journal-title":"J Cheminform"}],"container-title":["BMC Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-023-05485-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12859-023-05485-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12859-023-05485-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,18]],"date-time":"2023-11-18T21:38:10Z","timestamp":1700343490000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcbioinformatics.biomedcentral.com\/articles\/10.1186\/s12859-023-05485-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,19]]},"references-count":35,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["5485"],"URL":"https:\/\/doi.org\/10.1186\/s12859-023-05485-9","relation":{},"ISSN":["1471-2105"],"issn-type":[{"value":"1471-2105","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,19]]},"assertion":[{"value":"20 June 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 September 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 September 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"348"}}