{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T15:34:13Z","timestamp":1778168053899,"version":"3.51.4"},"reference-count":55,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T00:00:00Z","timestamp":1750982400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100009192","name":"Alberta Innovates","doi-asserted-by":"crossref","award":["#200700595"],"award-info":[{"award-number":["#200700595"]}],"id":[{"id":"10.13039\/501100009192","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100000024","name":"Canadian Institutes of Health Research","doi-asserted-by":"publisher","award":["PJT-173295"],"award-info":[{"award-number":["PJT-173295"]}],"id":[{"id":"10.13039\/501100000024","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Spontaneous preterm birth (sPTB) remains a significant global health challenge and a leading cause of neonatal mortality and morbidity. Despite advancements in neonatal care, the prediction of sPTB remains elusive, in part due to complex etiologies and heterogeneous patient populations. This study aimed to validate and extend information on gene expression biomarkers previously described for predicting sPTB using maternal whole blood from the All Our Families pregnancy cohort study based in Calgary,\u00a0Canada. The results of this study are two-fold: first, using additional replicates of maternal blood samples from the All Our Families cohort, we were unable to repeat the findings of a 2016 study which identified top maternal gene expression predictors for sPTB. Second, we conducted a secondary analysis of the original gene expression dataset from the 2016 study using five modelling approaches (random forest, elastic net regression, unregularized logistic regression, L2-regularized logistic regression, and multilayer perceptron neural network) followed by external validation using a pregnancy cohort based in Detroit, USA. The top performing model (random forest classification) suggested promising performance (area under the receiver operating curve, AUROC 0.99 in the training set), but performance was significantly degraded on the test set (AUROC 0.54) and further degraded in external validation (AUROC 0.50), suggesting poor generalizability, likely due to overfitting exacerbated by a low feature-to-noise ratio. Similar performance was observed in the other four learning models. Prediction was not improved when using higher complexity machine learning (e.g., neural network) approaches over traditional statistical learning (e.g., logistic regression). These findings underscore the challenges in translating biomarker discovery into clinically useful predictive models for sPTB. This study highlights the critical need for rigorous methodological safeguards and external validation in biomarker research. It also emphasizes the impact of data noise and overfitting on model performance, particularly in high-dimensional omics datasets. Future research should prioritize robust validation strategies and explore mechanistic insights to improve our understanding and prediction of sPTB.<\/jats:p>","DOI":"10.1371\/journal.pone.0310937","type":"journal-article","created":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T13:41:05Z","timestamp":1751031665000},"page":"e0310937","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":2,"title":["Machine learning for the prediction of spontaneous preterm birth using early second and third trimester maternal blood gene expression: A cautionary tale"],"prefix":"10.1371","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0876-5417","authenticated-orcid":true,"given":"Kylie K.","family":"Hornaday","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ty","family":"Werbicki","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Suzanne C.","family":"Tough","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Stephen L.","family":"Wood","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David W.","family":"Anderson","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Constance H.","family":"Li","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Donna M.","family":"Slater","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2025,6,27]]},"reference":[{"issue":"10409","key":"pone.0310937.ref001","doi-asserted-by":"crossref","first-page":"1261","DOI":"10.1016\/S0140-6736(23)00878-4","article-title":"National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis","volume":"402","author":"EO Ohuma","year":"2023","journal-title":"Lancet"},{"issue":"7","key":"pone.0310937.ref002","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1053\/j.semperi.2017.07.009","article-title":"Epidemiology of preterm birth","volume":"41","author":"SE Purisch","year":"2017","journal-title":"Semin Perinatol"},{"issue":"2","key":"pone.0310937.ref003","doi-asserted-by":"crossref","first-page":"106","DOI":"10.1016\/S2352-4642(21)00311-4","article-title":"Global, regional, and national causes of under-5 mortality in 2000-19: an updated systematic analysis with implications for the Sustainable Development Goals","volume":"6","author":"J Perin","year":"2022","journal-title":"Lancet Child Adolesc Health"},{"issue":"4","key":"pone.0310937.ref004","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0265853","article-title":"Is there a maternal blood biomarker that can predict spontaneous preterm birth prior to labour onset? A systematic review","volume":"17","author":"KK Hornaday","year":"2022","journal-title":"PLoS One"},{"issue":"2","key":"pone.0310937.ref005","doi-asserted-by":"crossref","first-page":"391","DOI":"10.1016\/j.clp.2024.02.011","article-title":"Predicting preterm birth using proteomics","volume":"51","author":"I Mari\u0107","year":"2024","journal-title":"Clin Perinatol"},{"issue":"3","key":"pone.0310937.ref006","doi-asserted-by":"crossref","first-page":"932","DOI":"10.1002\/ijgo.15514","article-title":"Prediction and prevention of preterm birth: quality assessment and systematic review of clinical practice guidelines using the AGREE II framework","volume":"166","author":"A Ramachandran","year":"2024","journal-title":"Int J Gynaecol Obstet"},{"issue":"1","key":"pone.0310937.ref007","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1111\/aogs.14475","article-title":"Reporting and risk of bias of prediction models based on machine learning methods in preterm birth: a systematic review","volume":"102","author":"Q Yang","year":"2023","journal-title":"Acta Obstet Gynecol Scand"},{"issue":"3","key":"pone.0310937.ref008","doi-asserted-by":"crossref","first-page":"396","DOI":"10.1097\/ALN.0000000000003871","article-title":"Statistical development and validation of clinical prediction models","volume":"135","author":"SJ Staffa","year":"2021","journal-title":"Anesthesiology"},{"key":"pone.0310937.ref009","doi-asserted-by":"crossref","first-page":"245","DOI":"10.1016\/j.jclinepi.2015.04.005","article-title":"Prediction models need appropriate internal, internal-external, and external validation","volume":"69","author":"EW Steyerberg","year":"2016","journal-title":"J Clin Epidemiol"},{"issue":"4","key":"pone.0310937.ref010","doi-asserted-by":"crossref","DOI":"10.2196\/33875","article-title":"Machine learning approach for preterm birth prediction using health records: systematic review","volume":"10","author":"Z Sharifi-Heris","year":"2022","journal-title":"JMIR Med Inform"},{"key":"pone.0310937.ref011","doi-asserted-by":"crossref","first-page":"1130139","DOI":"10.3389\/fendo.2023.1130139","article-title":"Machine learning applied in maternal and fetal health: a narrative review focused on pregnancy diseases and complications","volume":"14","author":"D Mennickent","year":"2023","journal-title":"Front Endocrinol (Lausanne)"},{"issue":"5","key":"pone.0310937.ref012","doi-asserted-by":"crossref","first-page":"623","DOI":"10.1097\/CCM.0000000000004246","article-title":"Development and reporting of prediction models: guidance for authors from editors of respiratory, sleep, and critical care journals","volume":"48","author":"DE Leisman","year":"2020","journal-title":"Crit Care Med"},{"key":"pone.0310937.ref013","article-title":"Evaluation of clinical prediction models (part 1): from development to external validation","volume":"384","author":"GS Collins","year":"2024","journal-title":"BMJ"},{"issue":"2","key":"pone.0310937.ref014","doi-asserted-by":"crossref","first-page":"396","DOI":"10.1097\/TP.0000000000003209","article-title":"External validation of the DynPG for kidney transplant recipients","volume":"105","author":"R Lenain","year":"2021","journal-title":"Transplantation"},{"issue":"3","key":"pone.0310937.ref015","doi-asserted-by":"crossref","DOI":"10.1002\/aet2.10887","article-title":"External validation of the ultrasound competency assessment tool","volume":"7","author":"FM Russell","year":"2023","journal-title":"AEM Educ Train"},{"key":"pone.0310937.ref016","doi-asserted-by":"crossref","first-page":"1006470","DOI":"10.3389\/fendo.2022.1006470","article-title":"External validation and clinical application of the predictive model for severe hypoglycemia","volume":"13","author":"J-S Yun","year":"2022","journal-title":"Front Endocrinol (Lausanne)"},{"key":"pone.0310937.ref017","article-title":"Performance of prediction models for nephropathy in people with type 2 diabetes: systematic review and external validation study","volume":"374","author":"RC Slieker","year":"2021","journal-title":"BMJ"},{"key":"pone.0310937.ref018","doi-asserted-by":"crossref","DOI":"10.1016\/j.ajog.2020.07.020","article-title":"First trimester preeclampsia screening and prediction","volume":"226","author":"P Chaemsaithong","year":"2022","journal-title":"Am J Obstet Gynecol"},{"issue":"1","key":"pone.0310937.ref019","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1111\/1471-0528.16379","article-title":"Predicting risk of postpartum haemorrhage: a systematic review","volume":"128","author":"C Neary","year":"2021","journal-title":"BJOG"},{"issue":"2","key":"pone.0310937.ref020","doi-asserted-by":"crossref","first-page":"441","DOI":"10.1016\/j.clp.2024.02.013","article-title":"Predicting spontaneous preterm birth using the immunome","volume":"51","author":"D Feyaerts","year":"2024","journal-title":"Clin Perinatol"},{"issue":"6","key":"pone.0310937.ref021","article-title":"Maternal whole blood gene expression at 18 and 28 weeks of gestation associated with spontaneous preterm birth in asymptomatic women","volume":"11","author":"YJ Heng","year":"2016","journal-title":"PLoS ONE"},{"issue":"12","key":"pone.0310937.ref022","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0242404","article-title":"Quality assessment of RNA in long-term storage: The All Our Families biorepository","volume":"15","author":"NL Stephenson","year":"2020","journal-title":"PLoS One"},{"issue":"6","key":"pone.0310937.ref023","doi-asserted-by":"crossref","first-page":"100323","DOI":"10.1016\/j.xcrm.2021.100323","article-title":"Crowdsourcing assessment of maternal blood multi-omics for predicting gestational age and preterm birth","volume":"2","author":"AL Tarca","year":"2021","journal-title":"Cell Rep Med"},{"key":"pone.0310937.ref024","volume-title":"Introduction to machine learning","author":"IEEE Xplore D, MIT Press P, Ebrary I","year":"2014"},{"key":"pone.0310937.ref025","volume-title":"Machine learning: a Bayesian and optimization perspective","author":"S Theodoridis","year":"2015"},{"issue":"12","key":"pone.0310937.ref026","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1145\/2500499","article-title":"Data science and prediction","volume":"56","author":"V Dhar","year":"2013","journal-title":"Commun ACM"},{"key":"pone.0310937.ref027","doi-asserted-by":"crossref","first-page":"100099","DOI":"10.1016\/j.crphys.2023.100099","article-title":"Machine learning and disease prediction in obstetrics","volume":"6","author":"Z Arain","year":"2023","journal-title":"Curr Res Physiol"},{"issue":"8","key":"pone.0310937.ref028","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0134619","article-title":"Inflammatory and angiogenic factors at mid-pregnancy are associated with spontaneous preterm birth in a cohort of Tanzanian women","volume":"10","author":"CR McDonald","year":"2015","journal-title":"PLoS ONE"},{"key":"pone.0310937.ref029","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1186\/1471-2393-10-87","article-title":"All Our Babies Cohort Study: recruitment of a cohort to predict women at risk of preterm birth through the examination of gene expression profiles and the environment","volume":"10","author":"SK Gracie","year":"2010","journal-title":"BMC Pregnancy Childbirth"},{"issue":"5","key":"pone.0310937.ref030","doi-asserted-by":"crossref","DOI":"10.1093\/ije\/dyw363","article-title":"Cohort profile: the All Our Babies pregnancy cohort (AOB)","volume":"46","author":"SC Tough","year":"2017","journal-title":"Int J Epidemiol"},{"issue":"1","key":"pone.0310937.ref031","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1093\/bioinformatics\/bts635","article-title":"STAR: ultrafast universal RNA-seq aligner","volume":"29","author":"A Dobin","year":"2013","journal-title":"Bioinformatics"},{"issue":"2","key":"pone.0310937.ref032","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1016\/j.tig.2005.12.005","article-title":"Reliability and reproducibility issues in DNA microarray measurements","volume":"22","author":"S Draghici","year":"2006","journal-title":"Trends Genet"},{"key":"pone.0310937.ref033","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1186\/1471-2105-3-22","article-title":"Microarray results: how accurate are they?","volume":"3","author":"R Kothapalli","year":"2002","journal-title":"BMC Bioinformatics"},{"key":"pone.0310937.ref034","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1016\/j.jclinepi.2019.02.004","article-title":"A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models","volume":"110","author":"E Christodoulou","year":"2019","journal-title":"J Clin Epidemiol"},{"key":"pone.0310937.ref035","first-page":"22","article-title":"Immunoglobulin and T cell receptor genes: IMGT(\u00ae) and the birth and rise of immunoinformatics","volume":"5","author":"M-P Lefranc","year":"2014","journal-title":"Front Immunol"},{"issue":"3","key":"pone.0310937.ref036","first-page":"2077","article-title":"Identification of G1 kinase activity for cdk6, a novel cyclin D partner","volume":"14","author":"M Meyerson","year":"1994","journal-title":"Mol Cell Biol"},{"issue":"1","key":"pone.0310937.ref037","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1186\/s40364-024-00557-1","article-title":"DeepMPTB: a vaginal microbiome-based deep neural network as artificial intelligence strategy for efficient preterm birth prediction","volume":"12","author":"O Chakoory","year":"2024","journal-title":"Biomark Res"},{"issue":"2","key":"pone.0310937.ref038","doi-asserted-by":"crossref","first-page":"2284384","DOI":"10.1080\/07853890.2023.2284384","article-title":"Association between interleukin-6 and preterm birth: a meta-analysis","volume":"55","author":"Y Chang","year":"2023","journal-title":"Ann Med"},{"issue":"2","key":"pone.0310937.ref039","doi-asserted-by":"crossref","first-page":"379","DOI":"10.1016\/j.clp.2024.02.008","article-title":"Predicting preterm birth using cell-free ribonucleic acid","volume":"51","author":"AD Cowan","year":"2024","journal-title":"Clin Perinatol"},{"issue":"7898","key":"pone.0310937.ref040","doi-asserted-by":"crossref","first-page":"689","DOI":"10.1038\/s41586-022-04410-z","article-title":"Early prediction of preeclampsia in pregnancy with cell-free RNA","volume":"602","author":"MN Moufarrej","year":"2022","journal-title":"Nature"},{"key":"pone.0310937.ref041","article-title":"A scoping review of preterm birth risk factors","volume":"41","author":"ID Tang","year":"2024","journal-title":"Am J Perinatol"},{"issue":"1","key":"pone.0310937.ref042","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1186\/s12950-023-00351-0","article-title":"Finding the priority and cluster of inflammatory biomarkers for infectious preterm birth: a systematic review","volume":"20","author":"J Li","year":"2023","journal-title":"J Inflamm (Lond)"},{"issue":"1","key":"pone.0310937.ref043","doi-asserted-by":"crossref","first-page":"494","DOI":"10.1186\/s12916-023-03171-4","article-title":"Risk factors for preterm birth: an umbrella review of meta-analyses of observational studies","volume":"21","author":"I Mitrogiannis","year":"2023","journal-title":"BMC Med"},{"issue":"13","key":"pone.0310937.ref044","doi-asserted-by":"crossref","first-page":"4360","DOI":"10.3390\/jcm12134360","article-title":"Risk scoring systems for preterm birth and their performance: a systematic review","volume":"12","author":"A Ferreira","year":"2023","journal-title":"J Clin Med"},{"issue":"2","key":"pone.0310937.ref045","doi-asserted-by":"crossref","first-page":"345","DOI":"10.1016\/j.clp.2024.02.003","article-title":"Computational approaches for connecting maternal stress to preterm birth","volume":"51","author":"A Mirzaei","year":"2024","journal-title":"Clin Perinatol"},{"key":"pone.0310937.ref046","doi-asserted-by":"crossref","first-page":"1981","DOI":"10.2147\/IJWH.S436624","article-title":"Preterm birth: screening and prediction","volume":"15","author":"L Creswell","year":"2023","journal-title":"Int J Womens Health"},{"issue":"12","key":"pone.0310937.ref047","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0311721","article-title":"Maternal cytokine profiles in second and early third trimester are not predictive of preterm birth","volume":"19","author":"KK Hornaday","year":"2024","journal-title":"PLoS One"},{"issue":"3","key":"pone.0310937.ref048","doi-asserted-by":"crossref","DOI":"10.1002\/prca.70004","article-title":"Application of proteomics in maternal and neonatal health: advancements and future directions","volume":"19","author":"R Elkahlout","year":"2025","journal-title":"Proteomics Clin Appl"},{"issue":"3","key":"pone.0310937.ref049","doi-asserted-by":"crossref","first-page":"271","DOI":"10.1080\/1354750X.2025.2475474","article-title":"A systematic review of first-trimester blood biomarkers associated with preterm prelabor rupture of the fetal membranes","volume":"30","author":"M Kirk","year":"2025","journal-title":"Biomarkers"},{"key":"pone.0310937.ref050","volume-title":"Methods of dealing with values below the limit of detection using SAS carry","author":"W Croghan","year":"2003"},{"key":"pone.0310937.ref051","doi-asserted-by":"crossref","DOI":"10.1002\/9781118646106.ch2","volume-title":"Foundations of imbalanced learning","author":"GM Weiss","year":"2013"},{"issue":"3","key":"pone.0310937.ref052","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1111\/1471-0528.17978","article-title":"Fetal fraction of cell-free DNA in the prediction of adverse pregnancy outcomes: a nationwide retrospective cohort study","volume":"132","author":"EC Becking","year":"2025","journal-title":"BJOG"},{"issue":"1","key":"pone.0310937.ref053","doi-asserted-by":"crossref","first-page":"763","DOI":"10.1186\/s12884-024-06933-x","article-title":"Development and validation of a spontaneous preterm birth risk prediction algorithm based on maternal bioinformatics: a single-center retrospective study","volume":"24","author":"Y Chen","year":"2024","journal-title":"BMC Pregnancy Childbirth"},{"issue":"1","key":"pone.0310937.ref054","doi-asserted-by":"crossref","first-page":"783","DOI":"10.1186\/s12884-024-06974-2","article-title":"Prediction of risk for early or very early preterm births using high-resolution urinary metabolomic profiling","volume":"24","author":"Y Zhang","year":"2024","journal-title":"BMC Pregnancy Childbirth"},{"issue":"17","key":"pone.0310937.ref055","doi-asserted-by":"crossref","first-page":"9536","DOI":"10.3390\/ijms25179536","article-title":"Maternal plasma miRNAs as early biomarkers of moderate-to-late-preterm birth","volume":"25","author":"F Ramzan","year":"2024","journal-title":"Int J Mol Sci"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0310937","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T13:41:19Z","timestamp":1751031679000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0310937"}},"subtitle":[],"editor":[{"given":"Addisalem Workie","family":"Demsash","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2025,6,27]]},"references-count":55,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2025,6,27]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0310937","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2024.09.11.24313497","asserted-by":"object"}]},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,27]]}}}