{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T17:00:32Z","timestamp":1780938032335,"version":"3.54.1"},"reference-count":55,"publisher":"Oxford University Press (OUP)","issue":"3","license":[{"start":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T00:00:00Z","timestamp":1748304000000},"content-version":"vor","delay-in-days":26,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["JUSRP123035"],"award-info":[{"award-number":["JUSRP123035"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,5,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>More and more recent studies highlight the crucial role of the human microbiome in maintaining health, while modern advancements in metagenomic sequencing technologies have been accumulating data that are associated with human diseases. Although metagenomic data offer rich, multifaceted information, including taxonomic and functional abundance profiles, their full potential remains underutilized, as most approaches rely only on one type of information to discover and understand their related correlations with respect to disease occurrences. To address this limitation, we propose a multistage fusion tabular transformer architecture (MSFT-Transformer), aiming to effectively integrate various types of high-dimensional tabular information extracted from metagenomic data. Its multistage fusion strategy consists of three modules: a fusion-aware feature extraction module in the early stage to improve the extracted information from inputs, an alignment-enhanced fusion module in the mid stage to enforce the retainment of desired information in cross-modal learning, and an integrated feature decision layer in the late stage to incorporate desired cross-modal information. We conduct extensive experiments to evaluate the performance of MSFT-Transformer over state-of-the-art models on five standard datasets. Our results indicate that MSFT-Transformer provides stable performance gains with reduced computational costs. An ablation study illustrates the contributions of all three models compared with a reference multistage fusion transformer without these novel strategies. The result analysis implies the significant potential of the proposed model in future disease prediction with metagenomic data.<\/jats:p>","DOI":"10.1093\/bib\/bbaf217","type":"journal-article","created":{"date-parts":[[2025,5,15]],"date-time":"2025-05-15T06:26:44Z","timestamp":1747290404000},"source":"Crossref","is-referenced-by-count":5,"title":["MSFT-transformer: a multistage fusion tabular transformer for disease prediction using metagenomic data"],"prefix":"10.1093","volume":"26","author":[{"given":"Ning","family":"Wang","sequence":"first","affiliation":[{"name":"School of Artificial Intelligence and Computer Science , Jiangnan University, Wuxi 214121, Jiangsu,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-8434-4307","authenticated-orcid":false,"given":"Minghui","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Computer Science , Jiangnan University, Wuxi 214121, Jiangsu,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenchao","family":"Gu","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Computer Science , Jiangnan University, Wuxi 214121, Jiangsu,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chenglong","family":"Dai","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Computer Science , Jiangnan University, Wuxi 214121, Jiangsu,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zongru","family":"Shao","sequence":"additional","affiliation":[{"name":"Silicon Austria Labs , Linz,","place":["Austria"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"K P","family":"Subbalakshmi","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering , Stevens Institute of Technology, Castle Point Terrace, Hoboken, NJ 07030,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2025,5,15]]},"reference":[{"key":"2025052711504569200_ref1","doi-asserted-by":"publisher","first-page":"242","DOI":"10.1159\/000481627","article-title":"Gut microbiota in health and disease","volume":"71","author":"Yamashiro","year":"2018","journal-title":"Ann Nutr Metab"},{"key":"2025052711504569200_ref2","doi-asserted-by":"publisher","first-page":"833","DOI":"10.1038\/nbt.3935","article-title":"Shotgun metagenomics, from sampling to analysis","volume":"35","author":"Quince","year":"2017","journal-title":"Nat Biotechnol"},{"key":"2025052711504569200_ref3","doi-asserted-by":"publisher","first-page":"481","DOI":"10.1038\/455481a","article-title":"Metagenomics","volume":"455","author":"Hugenholtz","year":"2008","journal-title":"Nature"},{"key":"2025052711504569200_ref4","doi-asserted-by":"crossref","first-page":"986","DOI":"10.1109\/TCBB.2022.3201295","article-title":"EnsDeepDP: an ensemble deep learning approach for disease prediction through metagenomics","volume":"20","author":"Shen","year":"2022","journal-title":"IEEE\/ACM Trans Comput Biol Bioinform"},{"key":"2025052711504569200_ref5","doi-asserted-by":"publisher","first-page":"669","DOI":"10.3390\/life12050669","article-title":"MegaD: deep learning for rapid and accurate disease status prediction of metagenomic samples","volume":"12","author":"Mreyoud","year":"2022","journal-title":"Life"},{"key":"2025052711504569200_ref6","doi-asserted-by":"publisher","first-page":"1902","DOI":"10.1093\/bib\/bbaa043","article-title":"Application of deep learning methods in biological networks","volume":"22","author":"Jin","year":"2021","journal-title":"Brief Bioinform"},{"key":"2025052711504569200_ref7","doi-asserted-by":"crossref","first-page":"6026","DOI":"10.1038\/s41598-020-63159-5","article-title":"DeepMicro: deep representation learning for disease prediction based on microbiome data","volume":"10","author":"Min","year":"2020","journal-title":"Sci Rep"},{"key":"2025052711504569200_ref8","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1038\/nature12198","article-title":"Gut metagenome in European women with normal, impaired and diabetic glucose control","volume":"498","author":"Karlsson","year":"2013","journal-title":"Nature"},{"key":"2025052711504569200_ref9","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1038\/nature13568","article-title":"Alterations of the human gut microbiome in liver cirrhosis","volume":"513","author":"Qin","year":"2014","journal-title":"Nature"},{"key":"2025052711504569200_ref10","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1038\/nature11450","article-title":"A metagenome-wide association study of gut microbiota in type 2 diabetes","volume":"490","author":"Qin","year":"2012","journal-title":"Nature"},{"key":"2025052711504569200_ref11","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1038\/nature08821","article-title":"A human gut microbial gene catalogue established by metagenomic sequencing","volume":"464","author":"Qin","year":"2010","journal-title":"Nature"},{"key":"2025052711504569200_ref12","doi-asserted-by":"publisher","first-page":"541","DOI":"10.1038\/nature12506","article-title":"Richness of human gut microbiome correlates with metabolic markers","volume":"500","author":"Le Chatelier","year":"2013","journal-title":"Nature"},{"key":"2025052711504569200_ref13","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s12859-020-03933-4","article-title":"MegaR: an interactive r package for rapid sample classification and phenotype prediction using metagenome profiles and machine learning","volume":"22","author":"Dhungel","year":"2021","journal-title":"BMC Bioinformatics"},{"key":"2025052711504569200_ref14","doi-asserted-by":"publisher","first-page":"2993","DOI":"10.1109\/JBHI.2020.2993761","article-title":"PopPhy-CNN: a phylogenetic tree embedded architecture for convolutional neural networks to predict host phenotype from metagenomic data","volume":"24","author":"Reiman","year":"2020","journal-title":"IEEE J Biomed Health Inform"},{"key":"2025052711504569200_ref15","doi-asserted-by":"publisher","DOI":"10.12688\/f1000research.27384.1","article-title":"Meta-Signer: metagenomic signature identifier based on rank aggregation of features","volume-title":"F1000Research","author":"Reiman"},{"key":"2025052711504569200_ref16","doi-asserted-by":"publisher","first-page":"4544","DOI":"10.1093\/bioinformatics\/btaa542","article-title":"TaxoNN: ensemble of neural networks on stratified microbiome data for disease prediction","volume":"36","author":"Sharma","year":"2020","journal-title":"Bioinformatics"},{"key":"2025052711504569200_ref17","doi-asserted-by":"publisher","first-page":"104081","DOI":"10.1016\/j.isci.2022.104081","article-title":"Human disease prediction from microbiome data by multiple feature fusion and deep learning","volume":"25","author":"Chen","year":"2022","journal-title":"Iscience"},{"key":"2025052711504569200_ref18","doi-asserted-by":"publisher","first-page":"2302076","DOI":"10.1080\/19490976.2024.2302076","article-title":"Incorporating metabolic activity, taxonomy and community structure to improve microbiome-based predictive models for host phenotype prediction","volume":"16","author":"Monshizadeh","year":"2024","journal-title":"Gut Microbes"},{"key":"2025052711504569200_ref19","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1093\/nar\/28.1.27","article-title":"KEGG: Kyoto encyclopedia of genes and genomes","volume":"28","author":"Kanehisa","year":"2000","journal-title":"Nucleic Acids Res"},{"key":"2025052711504569200_ref20","doi-asserted-by":"publisher","first-page":"258D","DOI":"10.1093\/nar\/gkh036","article-title":"The gene ontology (go) database and informatics resource","volume":"32","author":"Gene Ontology Consortium","year":"2004","journal-title":"Nucleic Acids Res"},{"key":"2025052711504569200_ref21","doi-asserted-by":"publisher","first-page":"D309","DOI":"10.1093\/nar\/gky1085","article-title":"eggNOG 5.0: a hierarchical, functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses","volume":"47","author":"Huerta-Cepas","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2025052711504569200_ref22","doi-asserted-by":"publisher","first-page":"711244","DOI":"10.3389\/fmicb.2021.711244","article-title":"Identification of microbiota biomarkers with orthologous gene annotation for type 2 diabetes","volume":"12","author":"Zhang","year":"2021","journal-title":"Front Microbiol"},{"key":"2025052711504569200_ref23","doi-asserted-by":"crossref","first-page":"1873","DOI":"10.1109\/BIBM52615.2021.9669492","article-title":"Discovering microbe functionality in human disease with a gene-ontology-aware model","volume-title":"2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","author":"Liu","year":"2021"},{"key":"2025052711504569200_ref24","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"2025052711504569200_ref25","article-title":"Multimodal learning with transformers: a survey","volume":"45","author":"Peng","year":"2023","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2025052711504569200_ref26","first-page":"14200","article-title":"Attention bottlenecks for multimodal fusion","volume":"34","author":"Nagrani","year":"2021","journal-title":"Advances in neural information processing systems"},{"key":"2025052711504569200_ref27","doi-asserted-by":"publisher","first-page":"1773","DOI":"10.1038\/s41591-022-01981-2","article-title":"Multimodal biomedical AI","volume":"28","author":"Acosta","year":"2022","journal-title":"Nat Med"},{"key":"2025052711504569200_ref28","doi-asserted-by":"publisher","first-page":"824","DOI":"10.1038\/s41598-022-04773-3","article-title":"Multimodal deep learning applied to classify healthy and disease states of human microbiome","volume":"12","author":"Lee","year":"2022","journal-title":"Sci Rep"},{"key":"2025052711504569200_ref29","doi-asserted-by":"publisher","first-page":"e1010050","DOI":"10.1371\/journal.pcbi.1010050","article-title":"Microbiome-based disease prediction with multimodal variational information bottlenecks","volume":"18","author":"Grazioli","year":"2022","journal-title":"PLoS Comput Biol"},{"key":"2025052711504569200_ref30","doi-asserted-by":"publisher","first-page":"902","DOI":"10.1038\/nmeth.3589","article-title":"Metaphlan2 for enhanced metagenomic taxonomic profiling","volume":"12","author":"Truong","year":"2015","journal-title":"Nat Methods"},{"key":"2025052711504569200_ref31","doi-asserted-by":"publisher","first-page":"e1004977","DOI":"10.1371\/journal.pcbi.1004977","article-title":"Machine learning meta-analysis of large metagenomic datasets: tools and biological insights","volume":"12","author":"Pasolli","year":"2016","journal-title":"PLoS Comput Biol"},{"key":"2025052711504569200_ref32","doi-asserted-by":"publisher","first-page":"i884","DOI":"10.1093\/bioinformatics\/bty560","article-title":"fastp: an ultra-fast all-in-one FASTQ preprocessor","volume":"34","author":"Chen","year":"2018","journal-title":"Bioinformatics"},{"key":"2025052711504569200_ref33","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1038\/nmeth.1923","article-title":"Fast gapped-read alignment with bowtie 2","volume":"9","author":"Langmead","year":"2012","journal-title":"Nat Methods"},{"key":"2025052711504569200_ref34","doi-asserted-by":"publisher","first-page":"698286","DOI":"10.3389\/fmicb.2021.698286","article-title":"DiTing: a pipeline to infer and compare biogeochemical pathways from metagenomic and metatranscriptomic data","volume":"12","author":"Xue","year":"2021","journal-title":"Front Microbiol"},{"key":"2025052711504569200_ref35","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-19753-9","article-title":"Nonalcoholic fatty liver disease is associated with dysbiosis independent of body mass index and insulin resistance","volume":"8","author":"Da Silva","year":"2018","journal-title":"Sci Rep"},{"key":"2025052711504569200_ref36","first-page":"18932","article-title":"Revisiting deep learning models for tabular data","volume":"34","author":"Gorishniy","year":"2021","journal-title":"Advances in Neural Information Processing Systems (NeurIPS), Virtual Event,"},{"key":"2025052711504569200_ref37","article-title":"BERT: pre-training of deep bidirectional transformers for language understanding","volume-title":"Proceedings of NAACL-HLT, Minneapolis, MN, USA","author":"Devlin","year":"2019"},{"key":"2025052711504569200_ref38","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1007\/978-981-99-8429-91","article-title":"Learning bottleneck transformer for event image-voxel feature fusion based classification","volume-title":"Chinese Conference on Pattern Recognition and Computer Vision (PRCV)","author":"Chengguo Yuan","year":"2023"},{"key":"2025052711504569200_ref39","doi-asserted-by":"publisher","first-page":"3374","DOI":"10.1109\/TMI.2023.3287256","article-title":"Shared-specific feature learning with bottleneck fusion transformer for multi-modal whole slide image analysis","volume":"42","author":"Wang","year":"2023","journal-title":"IEEE Trans Med Imaging"},{"key":"2025052711504569200_ref40","doi-asserted-by":"publisher","first-page":"1065","DOI":"10.1016\/j.neuroscience.2006.08.035","article-title":"Multisensory processing via early cortical stages: connections of the primary auditory cortical field with other sensory systems","volume":"143","author":"Budinger","year":"2006","journal-title":"Neuroscience"},{"key":"2025052711504569200_ref41","article-title":"On the benefits of early fusion in multimodal representation learning","volume-title":"NeurIPS Workshop SVRHM","author":"Barnum","year":"2020"},{"key":"2025052711504569200_ref42","doi-asserted-by":"crossref","first-page":"bbab569","DOI":"10.1093\/bib\/bbab569","article-title":"Multimodal deep learning for biomedical data fusion: a review","volume":"23","author":"Stahlschmidt","year":"2022","journal-title":"Brief Bioinform"},{"key":"2025052711504569200_ref43","doi-asserted-by":"publisher","first-page":"762","DOI":"10.1016\/j.neucom.2006.10.011","article-title":"Neural input selection\u2014a fast model-based approach","volume":"70","author":"Li","year":"2007","journal-title":"Neurocomputing"},{"key":"2025052711504569200_ref44","doi-asserted-by":"crossref","first-page":"3893","DOI":"10.1145\/3394171.3413581","article-title":"Cross-modal relation-aware networks for audio-visual event localization","volume-title":"Proceedings of the 28th ACM International Conference on Multimedia","author":"Xu","year":"2020"},{"key":"2025052711504569200_ref45","article-title":"Incorporating bert into neural machine translation","volume-title":"International Conference on Learning Representations (ICLR)","author":"Zhu","year":"2019"},{"key":"2025052711504569200_ref46","first-page":"347","article-title":"CrossViT: cross-attention multi-scale vision transformer for image classification","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV)","author":"Chen","year":"2021"},{"key":"2025052711504569200_ref47","volume-title":"Neural Networks: A Comprehensive Foundation","author":"Haykin","year":"1994"},{"key":"2025052711504569200_ref48","first-page":"278","article-title":"Random decision forests","volume-title":"Proceedings of 3rd International Conference on Document Analysis and Recognition (ICDAR)","author":"Ho","year":"1995"},{"key":"2025052711504569200_ref49","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/BF00994018","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach Learn"},{"key":"2025052711504569200_ref50","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"2025052711504569200_ref51","doi-asserted-by":"crossref","first-page":"2585","DOI":"10.1007\/s10115-021-01605-0","article-title":"Model complexity of deep learning: a survey","volume":"63","author":"Xia","year":"2021","journal-title":"Knowl Inf Syst"},{"key":"2025052711504569200_ref52","article-title":"An analysis of deep neural network models for practical applications","author":"Canziani","year":"2016"},{"key":"2025052711504569200_ref53","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","year":"2017","journal-title":"Advances in neural information processing systems (NeurIPS)"},{"key":"2025052711504569200_ref54","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1186\/s12915-023-01578-2","article-title":"Indoleacrylic acid produced by parabacteroides distasonis alleviates type 2 diabetes via activation of AhR to repair intestinal barrier","volume":"21","author":"Liu","year":"2023","journal-title":"BMC Biol"},{"key":"2025052711504569200_ref55","doi-asserted-by":"publisher","first-page":"110407","DOI":"10.1016\/j.ygeno.2022.110407","article-title":"Plasma and urinary extracellular vesicle micrornas and their related pathways in diabetic kidney disease","volume":"114","author":"Park","year":"2022","journal-title":"Genomics"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/3\/bbaf217\/63183274\/bbaf217.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/3\/bbaf217\/63183274\/bbaf217.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T15:51:00Z","timestamp":1748361060000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf217\/8131741"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,1]]},"references-count":55,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,5,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf217","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,5]]},"published":{"date-parts":[[2025,5,1]]},"article-number":"bbaf217"}}