{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T09:56:01Z","timestamp":1791194161681,"version":"4.1.0"},"posted":{"date-parts":[[2026,9,29]]},"group-title":"Bioinformatics","reference-count":51,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T00:00:00Z","timestamp":1790640000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"award":["GM145772"],"award-info":[{"award-number":["GM145772"]}],"id":[{"id":"https:\/\/ror.org\/01cwqze88","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,9,29]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Understanding the causal pathways linking the gut microbiome to downstream biomarkers and clinical outcomes is central to microbiome research. When the microbiome acts as the exposure, the high dimensionality and compositional structure of the data introduce unique methodological challenges for mediation analysis. To address these challenges, we propose a latent variable mediation framework that captures variation in microbial composition through a microbial balance, defined as the log-ratio between two unknown subsets of taxa. This balance serves as a latent scalar exposure and simplifies the estimation of the overall indirect effect at the community level, while simultaneously identifying specific taxa that contribute to the overall direct and indirect effects. We describe the model\u2019s estimation and inference, and illustrate the method using data from the Multiethnic Cohort\u2013Adiposity Phenotype Study, where we examine the mediation pathway from the gut microbiome through lipopolysaccharide-binding protein, a marker of metabolic endotoxemia, to percent liver fat.<\/jats:p>","DOI":"10.64898\/2026.09.24.754147","type":"posted-content","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:55:13Z","timestamp":1790751313000},"source":"Crossref","is-referenced-by-count":0,"title":["Mediation Analysis with Compositional Exposures"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6294-227X","authenticated-orcid":false,"given":"Jing","family":"Ma","sequence":"first","affiliation":[{"name":"Public Health Sciences Division, Fred Hutchinson Cancer Center"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Poorbita","family":"Kundu","sequence":"additional","affiliation":[{"name":"Public Health Sciences Division, Fred Hutchinson Cancer Center"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Timothy","family":"Randolph","sequence":"additional","affiliation":[{"name":"Clinical Research Division, Fred Hutchinson Cancer Center"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sandi","family":"Navarro","sequence":"additional","affiliation":[{"name":"Public Health Sciences Division, Fred Hutchinson Cancer Center"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Meredith","family":"Hullar","sequence":"additional","affiliation":[{"name":"Public Health Sciences Division, Fred Hutchinson Cancer Center"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026100502400962000_2026.09.24.754147v1.1","doi-asserted-by":"crossref","unstructured":"J Aitchison . The Statistical Analysis of Compositional Data. Chapman & Hall, Ltd., GBR, 1986. ISBN 0412280604.","DOI":"10.1007\/978-94-009-4109-0"},{"key":"2026100502400962000_2026.09.24.754147v1.2","first-page":"387","article-title":"A concise guide for the algebraic-geometric structure of the simplex, the sample space for compositional data analysis","volume":"2","year":"2002","journal-title":"Proceedings of IAMG"},{"issue":"5","key":"2026100502400962000_2026.09.24.754147v1.3","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1038\/s41575-020-0269-9","article-title":"Gut microbiota and human NAFLD: disentangling microbial signatures from metabolic disorders","volume":"17","year":"2020","journal-title":"Nature Reviews Gastroenterology & Hepatology"},{"key":"2026100502400962000_2026.09.24.754147v1.4","doi-asserted-by":"publisher","DOI":"10.2337\/db06-1491"},{"key":"2026100502400962000_2026.09.24.754147v1.5","doi-asserted-by":"publisher","DOI":"10.1007\/s13253-014-0180-3"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.6","first-page":"418","article-title":"Variable selection for sparse dirichlet-multinomial regression with an application to microbiome data analysis","volume":"7","year":"2013","journal-title":"The Annals of Applied Statistics"},{"key":"2026100502400962000_2026.09.24.754147v1.7","doi-asserted-by":"publisher","DOI":"10.1007\/s11004-005-7381-9"},{"key":"2026100502400962000_2026.09.24.754147v1.8","doi-asserted-by":"publisher","DOI":"10.1023\/A:1023818214614"},{"key":"2026100502400962000_2026.09.24.754147v1.9","doi-asserted-by":"publisher","DOI":"10.1038\/nature15766"},{"issue":"17","key":"2026100502400962000_2026.09.24.754147v1.10","doi-asserted-by":"crossref","first-page":"2999","DOI":"10.1002\/sim.9764","article-title":"A bayesian joint model for compositional mediation effect selection in microbiome data","volume":"42","year":"2023","journal-title":"Statistics in Medicine"},{"issue":"11","key":"2026100502400962000_2026.09.24.754147v1.11","doi-asserted-by":"crossref","first-page":"1897","DOI":"10.1016\/j.chom.2024.09.014","article-title":"Vaginal lactobacilli produce anti-inflammatory \u03b2-carboline compounds","volume":"32","year":"2024","journal-title":"Cell Host & Microbe"},{"key":"2026100502400962000_2026.09.24.754147v1.12","doi-asserted-by":"crossref","first-page":"2224","DOI":"10.3389\/fmicb.2017.02224","article-title":"Microbiome datasets are compositional: and this is not optional","volume":"8","year":"2017","journal-title":"Frontiers in Microbiology"},{"key":"2026100502400962000_2026.09.24.754147v1.13","doi-asserted-by":"crossref","first-page":"100017","DOI":"10.1016\/j.acags.2019.100017","article-title":"Amalgamations are valid in compositional data analysis, can be used in agglomerative clustering, and their logratios have an inverse transformation","volume":"5","year":"2020","journal-title":"Applied Computing and Geosciences"},{"key":"2026100502400962000_2026.09.24.754147v1.14","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/57.1.97"},{"key":"2026100502400962000_2026.09.24.754147v1.15","doi-asserted-by":"publisher","DOI":"10.1186\/s13059-023-02902-3"},{"key":"2026100502400962000_2026.09.24.754147v1.16","unstructured":"Lu Huang . Bayesian Balance Regression and Mediation Analysis for Microbiome Compositional Data. PhD thesis, University of Pennsylvania, 2020."},{"key":"2026100502400962000_2026.09.24.754147v1.17","doi-asserted-by":"publisher","DOI":"10.1080\/19490976.2021.1965463"},{"key":"2026100502400962000_2026.09.24.754147v1.18","doi-asserted-by":"publisher","DOI":"10.1037\/a0020761"},{"key":"2026100502400962000_2026.09.24.754147v1.19","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2016.05.041"},{"key":"2026100502400962000_2026.09.24.754147v1.20","article-title":"Causal inference with high-dimensional treatments","year":"2026","journal-title":"arXiv preprint"},{"issue":"6","key":"2026100502400962000_2026.09.24.754147v1.21","first-page":"e01892","article-title":"Identification of 1,400 metabolites as mediators of obesity in 473 gut microbiota taxa: a mediation mendelian randomization study","volume":"13","year":"2025","journal-title":"Microbiology Spectrum"},{"key":"2026100502400962000_2026.09.24.754147v1.22","doi-asserted-by":"crossref","unstructured":"Ying Li , Xin Wang , Zitong Zhang , Lei Shi , Liang Cheng , and Xue Zhang . Effect of the gut microbiome, plasma metabolome, peripheral cells, and inflammatory cytokines on obesity: a bidirectional two-sample mendelian randomization study and mediation analysis. Frontiers in Immunology, 15:1348347, 2024.","DOI":"10.3389\/fimmu.2024.1348347"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.23","first-page":"bbae653","article-title":"Phyimpute and unifracimpute: two imputation approaches incorporating phylogeny information for microbial count data","volume":"26","year":"2025","journal-title":"Briefings in Bioinformatics"},{"key":"2026100502400962000_2026.09.24.754147v1.24","doi-asserted-by":"crossref","unstructured":"Josep Antoni Mart\u00edn-Fern\u00e1ndez , Javier Palarea-Albaladejo , and Ricardo Antonio Olea . Dealing with zeros. Compositional Data Analysis, pages 43\u201358, 2011.","DOI":"10.1002\/9781119976462.ch4"},{"key":"2026100502400962000_2026.09.24.754147v1.25","doi-asserted-by":"publisher","DOI":"10.1177\/1471082X14535524"},{"issue":"3","key":"2026100502400962000_2026.09.24.754147v1.26","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/s11004-017-9712-z","article-title":"Advances in principal balances for compositional data","volume":"50","year":"2018","journal-title":"Mathematical Geosciences"},{"key":"2026100502400962000_2026.09.24.754147v1.27","doi-asserted-by":"crossref","unstructured":"Balapuwaduge Isuru Layan Madusanka Mendis , L Sarvananda , Thilini N Jayasinghe , Iyanthimala Harshini Rajapakse , and Arosha Sampath Dissanayake . Mechanisms and key mediators of gut microbiota and type 2 diabetes mellitus: A comprehensive overview. Medicine in Microecology, page 100144, 2025.","DOI":"10.1016\/j.medmic.2025.100144"},{"key":"2026100502400962000_2026.09.24.754147v1.28","doi-asserted-by":"publisher","DOI":"10.1038\/ijo.2011.256"},{"key":"2026100502400962000_2026.09.24.754147v1.29","article-title":"Semiparametric causal sufficient dimension reduction of multidimensional treatments","year":"2017","journal-title":"arXiv preprint"},{"key":"2026100502400962000_2026.09.24.754147v1.30","unstructured":"Vera Pawlowsky-Glahn , Juan Jos\u00e9 Egozcue Raimon Tolosana Delgado , et al. Principal balances. In Proceedings of the 4th International Workshop on Compositional Data Analysis. Universitat de Girona. Departament d?Inform\u00e0tica i Matem\u00e0tica Aplicada, 2011."},{"key":"2026100502400962000_2026.09.24.754147v1.31","article-title":"Uncertainty quantification and posterior sampling for network reconstruction","year":"2025","journal-title":"arXiv preprint"},{"key":"2026100502400962000_2026.09.24.754147v1.32","doi-asserted-by":"publisher","DOI":"10.1128\/mSystems.00053-18"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.33","doi-asserted-by":"crossref","first-page":"3294","DOI":"10.1038\/s41467-018-05470-4","article-title":"Microbial tryptophan catabolites in health and disease","volume":"9","year":"2018","journal-title":"Nature Communications"},{"key":"2026100502400962000_2026.09.24.754147v1.34","doi-asserted-by":"publisher","DOI":"10.1093\/biomet\/70.1.41"},{"key":"2026100502400962000_2026.09.24.754147v1.35","doi-asserted-by":"publisher","DOI":"10.2307\/2287653"},{"key":"2026100502400962000_2026.09.24.754147v1.36","article-title":"Bayesian multinomial logistic normal models through marginally latent matrix-t processes","year":"2019","journal-title":"arXiv preprint"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.37","first-page":"661","article-title":"Compositional mediation analysis for microbiome studies","volume":"13","year":"2019","journal-title":"The Annals of Applied Statistics"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.38","first-page":"16","article-title":"A compositional mediation model for a binary outcome: application to microbiome studies","volume":"38","year":"2022","journal-title":"Bioinformatics"},{"issue":"27","key":"2026100502400962000_2026.09.24.754147v1.39","doi-asserted-by":"crossref","first-page":"6038","DOI":"10.1002\/sim.9168","article-title":"Bayesian hierarchical models for high-dimensional mediation analysis with coordinated selection of correlated mediators","volume":"40","year":"2021","journal-title":"Statistics in Medicine"},{"issue":"15","key":"2026100502400962000_2026.09.24.754147v1.40","doi-asserted-by":"crossref","first-page":"2296","DOI":"10.3390\/cells11152296","article-title":"Gut microbiota-derived tryptophan metabolites maintain gut and systemic homeostasis","volume":"11","year":"2022","journal-title":"Cells"},{"issue":"4","key":"2026100502400962000_2026.09.24.754147v1.41","first-page":"1701","article-title":"Markov chains for exploring posterior distributions","volume":"22","year":"1994","journal-title":"The Annals of Statistics"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.42","first-page":"95","article-title":"Mediation analysis with multiple mediators","volume":"2","year":"2014","journal-title":"Epidemiologic Methods"},{"key":"2026100502400962000_2026.09.24.754147v1.43","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-publhealth-032315-021402"},{"key":"2026100502400962000_2026.09.24.754147v1.44","doi-asserted-by":"publisher","DOI":"10.1097\/EDE.0000000000000034"},{"key":"2026100502400962000_2026.09.24.754147v1.45","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btz565"},{"key":"2026100502400962000_2026.09.24.754147v1.46","doi-asserted-by":"publisher","DOI":"10.1002\/sim.8020"},{"key":"2026100502400962000_2026.09.24.754147v1.47","first-page":"1","article-title":"Guild-based analysis for understanding gut microbiome in human health and diseases","volume":"13","year":"2021","journal-title":"Genome Medicine"},{"issue":"1","key":"2026100502400962000_2026.09.24.754147v1.48","doi-asserted-by":"crossref","first-page":"5110276","DOI":"10.1155\/2021\/5110276","article-title":"Gut microbiota and type 2 diabetes mellitus: association, mechanism, and translational applications","volume":"2021","year":"2021","journal-title":"Mediators of Inflammation"},{"key":"2026100502400962000_2026.09.24.754147v1.49","doi-asserted-by":"publisher","DOI":"10.1002\/sim.10215"},{"key":"2026100502400962000_2026.09.24.754147v1.50","doi-asserted-by":"publisher","DOI":"10.1002\/hbm.25800"},{"issue":"10","key":"2026100502400962000_2026.09.24.754147v1.51","doi-asserted-by":"crossref","first-page":"e1013124","DOI":"10.1371\/journal.pcbi.1013124","article-title":"Bmdd: A probabilistic framework for accurate imputation of zero-inflated microbiome sequencing data","volume":"21","year":"2025","journal-title":"PLOS Computational Biology"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.24.754147","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T09:40:20Z","timestamp":1791193220000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.09.24.754147"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":51,"URL":"https:\/\/doi.org\/10.64898\/2026.09.24.754147","relation":{},"subject":[],"published":{"date-parts":[[2026,9,29]]},"subtype":"preprint"}}