{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T18:51:24Z","timestamp":1791226284717,"version":"4.1.0"},"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T00:00:00Z","timestamp":1791158400000},"content-version":"am","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Institute of Health","award":["KLTR001429-09"],"award-info":[{"award-number":["KLTR001429-09"]}]},{"DOI":"10.13039\/100000071","name":"National Institute of Child Health and Human Development","doi-asserted-by":"publisher","award":["DP1HD115449"],"award-info":[{"award-number":["DP1HD115449"]}],"id":[{"id":"10.13039\/100000071","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100023581","name":"National Science Foundation Graduate Research Fellowship Program","doi-asserted-by":"publisher","award":["DGE-2236414"],"award-info":[{"award-number":["DGE-2236414"]}],"id":[{"id":"10.13039\/100023581","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>In this study we utilize machine learning techniques to create predictive models and determine key bacterial interactions for the diagnosis of Bacterial vaginosis. Bacterial vaginosis (BV) is a common vaginal syndrome affecting reproductive-age women globally. It is associated with various adverse obstetric and gynecological out-comes including increased risk of sexually transmitted infections, HIV, cervical cancer, and pre-term birth. While it is known that BV is caused by a shift in abundance between Lactobacilli and anaerobic bacteria, it is unknown how gradual shifts in that balance lead towards BV status. Here we perform a rigorous comparison of machine learning architectures and feature selection methods used to train models on 16s rRNA data of patients presenting with BV. Using the highest-performing models, we employ explainable AI methods to determine the most important bacteria for BV diagnosis. Furthermore, we implement Voronoi-based decision boundaries to show how the relative abundances between pairs of these bacteria results in BV positive or BV negative outcomes. Results: We find that support vector machine and random forest models in combination with feature selection predict BV diagnosis with the most balanced accuracy. Using those models, we identify four Lactobacilli spp and six anaerobes to be key in to be key to the diagnosis of BV. The determination of key bacteria can inform BV diagnostics and pathogenesis research to species that have previously eluded scientific focus. Additionally, decision boundary plots offer a diagnostic point of reference for how the relative abundances of key vaginal flora are indicative of BV outcomes.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1014767","type":"journal-article","created":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T17:59:23Z","timestamp":1791223163000},"page":"e1014767","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Machine learning and Voronoi-based decision boundaries for Bacterial vaginosis to determine population- specific microbial interactions"],"prefix":"10.1371","volume":"22","author":[{"given":"Cameron G.","family":"Celeste","sequence":"first","affiliation":[{"name":"University of Florida","place":["Gainesville, Florida, United States of America"],"department":["J. Crayton Pruitt Department of Biomedical Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carleigh C.","family":"Sokolik","sequence":"additional","affiliation":[{"name":"University of Florida","place":["Gainesville, Florida, United States of America"],"department":["J. Crayton Pruitt Department of Biomedical Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wambui","family":"Gachunga","sequence":"additional","affiliation":[{"name":"University of Florida","place":["Gainesville, Florida, United States of America"],"department":["J. Crayton Pruitt Department of Biomedical Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2445-0765","authenticated-orcid":true,"given":"Ivana K.","family":"Parker","sequence":"additional","affiliation":[{"name":"University of Florida","place":["Gainesville, Florida, United States of America"],"department":["J. Crayton Pruitt Department of Biomedical Engineering"]},{"name":"University of Florida","place":["Gainesville, Florida, United States of America"],"department":["Emerging Pathogens Institute"]},{"name":"University of Florida","place":["Gainesville, Florida, United States of America"],"department":["Department of African American Studies"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,5]]},"reference":[{"key":"pcbi.1014767.ref001","doi-asserted-by":"crossref","first-page":"631972","DOI":"10.3389\/fcimb.2021.631972","article-title":"The female vaginal microbiome in health and bacterial vaginosis","volume":"11","author":"X Chen","year":"2021","journal-title":"Front Cell Infect Microbiol"},{"issue":"5","key":"pcbi.1014767.ref002","doi-asserted-by":"crossref","first-page":"663","DOI":"10.1086\/367658","article-title":"Bacterial vaginosis is a strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection","volume":"36","author":"HC Wiesenfeld","year":"2003","journal-title":"Clin Infect Dis"},{"issue":"11","key":"pcbi.1014767.ref003","doi-asserted-by":"crossref","first-page":"1723","DOI":"10.1093\/infdis\/jiu330","article-title":"Interplay between the temporal dynamics of the vaginal microbiota and human papillomavirus detection","volume":"210","author":"RM Brotman","year":"2014","journal-title":"J Infect Dis"},{"issue":"3","key":"pcbi.1014767.ref004","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1016\/j.ajog.2020.10.019","article-title":"Bacterial vaginosis and its association with infertility, endometritis, and pelvic inflammatory disease","volume":"224","author":"J Ravel","year":"2021","journal-title":"Am J Obstet Gynecol"},{"issue":"3","key":"pcbi.1014767.ref005","doi-asserted-by":"crossref","first-page":"319","DOI":"10.1086\/375819","article-title":"Association between acquisition of herpes simplex virus type 2 in women and bacterial vaginosis","volume":"37","author":"TL Cherpes","year":"2003","journal-title":"Clin Infect Dis"},{"issue":"6","key":"pcbi.1014767.ref006","doi-asserted-by":"crossref","first-page":"585","DOI":"10.1093\/aje\/kwi243","article-title":"A cluster analysis of bacterial vaginosis-associated microflora and pelvic inflammatory disease","volume":"162","author":"RB Ness","year":"2005","journal-title":"Am J Epidemiol"},{"issue":"12","key":"pcbi.1014767.ref007","doi-asserted-by":"crossref","first-page":"1493","DOI":"10.1097\/QAD.0b013e3283021a37","article-title":"Bacterial vaginosis and HIV acquisition: A meta-analysis of published studies","volume":"22","author":"J Atashili","year":"2008","journal-title":"AIDS"},{"issue":"1","key":"pcbi.1014767.ref008","first-page":"14","article-title":"Nonspecific vaginitis. Diagnostic criteria and microbial and epidemiologic associations","volume":"74","author":"R Amsel","year":"1983","journal-title":"Am J Med"},{"issue":"2","key":"pcbi.1014767.ref009","doi-asserted-by":"crossref","first-page":"297","DOI":"10.1128\/jcm.29.2.297-301.1991","article-title":"Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation","volume":"29","author":"RP Nugent","year":"1991","journal-title":"J Clin Microbiol"},{"key":"pcbi.1014767.ref010","doi-asserted-by":"crossref","first-page":"2272","DOI":"10.1099\/mic.0.081034-0","article-title":"Differences in vaginal microbiome in African American women versus women of European ancestry","volume":"160","author":"JM Fettweis","year":"2014","journal-title":"Microbiology (Reading)"},{"key":"pcbi.1014767.ref011","doi-asserted-by":"crossref","first-page":"864","DOI":"10.1097\/OLQ.0b013e318074e565","article-title":"The prevalence of bacterial vaginosis in the United States, 2001\u20132004; associations with symptoms, sexual behaviors, and reproductive health","volume":"34","author":"EH Koumans","year":"2007","journal-title":"Sex Transm Dis"},{"key":"pcbi.1014767.ref012","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1007\/s11908-020-00740-z","article-title":"Asymptomatic bacterial vaginosis: To treat or not to treat?","volume":"22","author":"CA Muzny","year":"2020","journal-title":"Curr Infect Dis Rep"},{"issue":"6","key":"pcbi.1014767.ref013","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0037818","article-title":"Bacterial communities in women with bacterial vaginosis: High resolution phylogenetic analyses reveal relationships of microbiota to clinical criteria","volume":"7","author":"S Srinivasan","year":"2012","journal-title":"PLoS One"},{"key":"pcbi.1014767.ref014","doi-asserted-by":"crossref","first-page":"4680","DOI":"10.1073\/pnas.1002611107","article-title":"Vaginal microbiome of reproductive-age women","author":"J Ravel","year":"2011","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pcbi.1014767.ref015","doi-asserted-by":"crossref","unstructured":"Baker YS, Agrawal R, Foster JA, Beck D, Dozier G. Detecting bacterial vaginosis using machine learning. Proceedings of the 2014 ACM Southeast Regional Conference. Kennesaw, Georgia: ACM; 2014. p. 1\u20134. https:\/\/doi.org\/10.1145\/2638404.2638521","DOI":"10.1145\/2638404.2638521"},{"key":"pcbi.1014767.ref016","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1186\/s13040-015-0055-3","article-title":"Machine learning classifiers provide insight into the relationship between microbial communities and bacterial vaginosis","volume":"8","author":"D Beck","year":"2015","journal-title":"BioData Min"},{"issue":"1","key":"pcbi.1014767.ref017","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1038\/s41746-023-00953-1","article-title":"Ethnic disparity in diagnosing asymptomatic bacterial vaginosis using machine learning","volume":"6","author":"C Celeste","year":"2023","journal-title":"npj Digit Med"},{"key":"pcbi.1014767.ref018","doi-asserted-by":"crossref","first-page":"101805","DOI":"10.1016\/j.inffus.2023.101805","article-title":"Explainable Artificial Intelligence (XAI): What we know and what is left to attain Trustworthy Artificial Intelligence","volume":"99","author":"S Ali","year":"2023","journal-title":"Inf Fusion"},{"issue":"3","key":"pcbi.1014767.ref019","doi-asserted-by":"crossref","first-page":"525","DOI":"10.3390\/make3030027","article-title":"Deterministic local interpretable model-agnostic explanations for stable explainability","volume":"3","author":"MR Zafar","year":"2021","journal-title":"MAKE"},{"issue":"10","key":"pcbi.1014767.ref020","doi-asserted-by":"crossref","first-page":"1340","DOI":"10.1093\/bioinformatics\/btq134","article-title":"Permutation importance: A corrected feature importance measure","volume":"26","author":"A Altmann","year":"2010","journal-title":"Bioinformatics"},{"key":"pcbi.1014767.ref021","first-page":"2579","article-title":"Visualizing high-dimensional data using t-SNE","volume":"9","author":"LJP van der Maaten","year":"2008","journal-title":"J Mach Learn Res"},{"key":"pcbi.1014767.ref022","article-title":"Mixtures of probabilistic principal component analysers","author":"ME Tipping","journal-title":"Neural Comput"},{"key":"pcbi.1014767.ref023","article-title":"UMAP: Uniform manifold approximation and projection for dimension reduction","author":"L McInnes","year":"2020"},{"issue":"1","key":"pcbi.1014767.ref024","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/s10618-013-0342-x","article-title":"Visualizing multi-dimensional decision boundaries in 2D","volume":"29","author":"MA Migut","year":"2013","journal-title":"Data Min Knowl Disc"},{"key":"pcbi.1014767.ref025","doi-asserted-by":"crossref","unstructured":"Muthukrishnan R, Rohini R. LASSO: A feature selection technique in predictive modeling for machine learning. 2016 IEEE International Conference on Advances in Computer Applications (ICACA); 2016. p. 18\u201320. https:\/\/doi.org\/10.1109\/icaca.2016.7887916","DOI":"10.1109\/ICACA.2016.7887916"},{"issue":"18","key":"pcbi.1014767.ref026","doi-asserted-by":"crossref","first-page":"1899","DOI":"10.1056\/NEJMoa043802","article-title":"Molecular identification of bacteria associated with bacterial vaginosis","volume":"353","author":"DN Fredricks","year":"2005","journal-title":"N Engl J Med"},{"key":"pcbi.1014767.ref027","doi-asserted-by":"crossref","first-page":"488","DOI":"10.1186\/1471-2164-11-488","article-title":"Molecular analysis of the diversity of vaginal microbiota associated with bacterial vaginosis","volume":"11","author":"Z Ling","year":"2010","journal-title":"BMC Genom"},{"issue":"5","key":"pcbi.1014767.ref028","doi-asserted-by":"crossref","first-page":"962","DOI":"10.1016\/0002-9378(55)90095-8","article-title":"Haemophilus vaginalis vaginitis: A newly defined specific infection previously classified non-specific vaginitis","volume":"69","author":"HL Gardner","year":"1955","journal-title":"Am J Obstet Gynecol"},{"issue":"10","key":"pcbi.1014767.ref029","doi-asserted-by":"crossref","first-page":"3270","DOI":"10.1128\/JCM.01272-07","article-title":"Targeted PCR for detection of vaginal bacteria associated with bacterial vaginosis","volume":"45","author":"DN Fredricks","year":"2007","journal-title":"J Clin Microbiol"},{"issue":"5","key":"pcbi.1014767.ref030","doi-asserted-by":"crossref","first-page":"1812","DOI":"10.1128\/JCM.00851-09","article-title":"Quantitative PCR assessments of bacterial species in women with and without bacterial vaginosis","volume":"48","author":"M Zozaya-Hinchliffe","year":"2010","journal-title":"J Clin Microbiol"},{"issue":"10","key":"pcbi.1014767.ref031","doi-asserted-by":"crossref","first-page":"1682","DOI":"10.1093\/infdis\/jiu303","article-title":"Gardnerella vaginalis does not always cause bacterial vaginosis","volume":"210","author":"RJ Hickey","year":"2014","journal-title":"J Infect Dis"},{"issue":"7","key":"pcbi.1014767.ref032","doi-asserted-by":"crossref","first-page":"1099","DOI":"10.1093\/infdis\/jiy704","article-title":"Gardnerella vaginalis and Prevotella bivia trigger distinct and overlapping phenotypes in a mouse model of bacterial vaginosis","volume":"220","author":"NM Gilbert","year":"2019","journal-title":"J Infect Dis"},{"key":"pcbi.1014767.ref033","doi-asserted-by":"crossref","first-page":"759697","DOI":"10.3389\/fcimb.2021.759697","article-title":"Immunometabolic analysis of Mobiluncus mulieris and Eggerthella sp. reveals novel insights into their pathogenic contributions to the hallmarks of bacterial vaginosis","volume":"11","author":"R McKenzie","year":"2021","journal-title":"Front Cell Infect Microbiol"},{"issue":"10","key":"pcbi.1014767.ref034","doi-asserted-by":"crossref","first-page":"2130","DOI":"10.3390\/v14102130","article-title":"Is the physiological composition of the vaginal microbiome altered in high-risk HPV infection of the uterine cervix?","volume":"14","author":"T Rokos","year":"2022","journal-title":"Viruses"},{"key":"pcbi.1014767.ref035","doi-asserted-by":"crossref","first-page":"709","DOI":"10.3390\/antibiotics9100709","article-title":"Antibiotic resistance of human periodontal pathogen Parvimonas micra over 10 years","volume":"9","author":"TE Rams","year":"2020","journal-title":"Antibiotics"},{"issue":"2","key":"pcbi.1014767.ref036","doi-asserted-by":"crossref","first-page":"310","DOI":"10.1016\/j.ajog.2021.11.1312","article-title":"Co-infection of bacterial vaginosis and mycoplasma genitalium in pregnant women","volume":"226","author":"S Cotter","year":"2022","journal-title":"Am J Obstet Gynecol"},{"key":"pcbi.1014767.ref037","doi-asserted-by":"crossref","first-page":"16445","DOI":"10.3390\/microorganisms10081645","article-title":"Vaginal Leptothrix: An innocent bystander?","volume":"10","author":"P Vieira-Baptista","year":"2022","journal-title":"Microorganisms"},{"issue":"1","key":"pcbi.1014767.ref038","doi-asserted-by":"crossref","first-page":"111","DOI":"10.1007\/s15010-018-1175-8","article-title":"Emergent genital infection by Leptotrichia trevisanii","volume":"47","author":"JC Mora-Palma","year":"2019","journal-title":"Infection"},{"issue":"11","key":"pcbi.1014767.ref039","doi-asserted-by":"crossref","DOI":"10.1172\/JCI184322","article-title":"Social, microbial, and immune factors linking bacterial vaginosis and infectious diseases","volume":"135","author":"NM Gilbert","year":"2025","journal-title":"J Clin Invest"},{"key":"pcbi.1014767.ref040","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1038\/s44294-025-00092-w","article-title":"Population-level predictive variation in machine learning diagnosis of symptomatic bacterial vaginosis","volume":"3","author":"DP Ojo","year":"2025","journal-title":"npj Womens Health"},{"issue":"2","key":"pcbi.1014767.ref041","doi-asserted-by":"crossref","first-page":"2981","DOI":"10.1080\/21655979.2022.2027059","article-title":"Log (Lactobacillus crispatus\/ Gardnerella vaginalis): A new indicator of diagnosing bacterial vaginosis","volume":"13","author":"T Deng","year":"2022","journal-title":"Bioengineered"},{"key":"pcbi.1014767.ref042","unstructured":"Bergstra J, Yamins D, Cox DD. Making a science of model search: Hyperparameter optimization in hundreds of dimensions for vision architectures. ICML 2013 I-115 to I\u201323; 2013."},{"issue":"1","key":"pcbi.1014767.ref043","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1186\/s40168-023-01692-x","article-title":"Integrating compositional and functional content to describe vaginal microbiomes in health and disease","volume":"11","author":"JB Holm","year":"2023","journal-title":"Microbiome"},{"key":"pcbi.1014767.ref044","author":"MA Klebanoff"},{"issue":"8","key":"pcbi.1014767.ref045","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pbio.1001377","article-title":"The Human Microbiome Project: A community resource for the healthy human microbiome","volume":"10","author":"D Gevers","year":"2012","journal-title":"PLoS Biol"}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1014767","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T17:59:30Z","timestamp":1791223170000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1014767"}},"subtitle":[],"editor":[{"given":"Rajib","family":"Saha","sequence":"first","affiliation":[{"name":"University of Nebraska-Lincoln","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,5]]},"references-count":45,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,5]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1014767","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,5]]}}}