Dataset Information

DeepMPTB: a vaginal microbiome-based deep neural network as artificial intelligence strategy for efficient preterm birth prediction.

ABSTRACT: In recent decades, preterm birth (PTB) has become a significant research focus in the healthcare field, as it is a leading cause of neonatal mortality worldwide. Using five independent study cohorts including 1290 vaginal samples from 561 pregnant women who delivered at term (n = 1029) or prematurely (n = 261), we analysed vaginal metagenomics data for precise microbiome structure characterization. Then, a deep neural network (DNN) was trained to predict term birth (TB) and PTB with an accuracy of 84.10% and an area under the receiver operating characteristic curve (AUROC) of 0.875 ± 0.11. During a benchmarking process, we demonstrated that our DL model outperformed seven currently used machine learning algorithms. Finally, our results indicate that overall diversity of the vaginal microbiota should be taken in account to predict PTB and not specific species. This artificial-intelligence based strategy should be highly helpful for clinicians in predicting preterm birth risk, allowing personalized assistance to address various health issues. DeepMPTB is open source and free for academic use. It is licensed under a GNU Affero General Public License 3.0 and is available at https://deepmptb.streamlit.app/ . Source code is available at https://github.com/oschakoory/DeepMPTB and can be easily installed using Docker ( https://www.docker.com/ ).

SUBMITTER: Chakoory O

PROVIDER: S-EPMC10865581 | biostudies-literature | 2024 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

DeepMPTB: a vaginal microbiome-based deep neural network as artificial intelligence strategy for efficient preterm birth prediction.

Chakoory Oshma O Barra Vincent V Rochette Emmanuelle E Blanchon Loïc L Sapin Vincent V Merlin Etienne E Pons Maguelonne M Gallot Denis D Comtet-Marre Sophie S Peyret Pierre P

Biomarker research 20240214 1

In recent decades, preterm birth (PTB) has become a significant research focus in the healthcare field, as it is a leading cause of neonatal mortality worldwide. Using five independent study cohorts including 1290 vaginal samples from 561 pregnant women who delivered at term (n = 1029) or prematurely (n = 261), we analysed vaginal metagenomics data for precise microbiome structure characterization. Then, a deep neural network (DNN) was trained to predict term birth (TB) and PTB with an accuracy ...[more]

PMID: 38355595

Similar Datasets

Project description:Previous studies have investigated the associations between the vaginal microbiome and preterm birth, with the aim of determining whether differences in community patterns meaningfully alter risk and could therefore be the target of intervention. We report on vaginal microbial analysis of a nested case-control subset of the Pregnancy, Infection, and Nutrition (PIN) Study, including 464 White women (375 term birth and 89 spontaneous preterm birth, sPTB) and 360 Black women (276 term birth and 84 sPTB). We found that the microbiome of Black women has higher alpha-diversity, higher abundance of Lactobacillus iners, and lower abundance of Lactobacillus crispatus. However, among women who douche, there were no significant differences in microbiome by race. The sPTB-associated microbiome exhibited a lower abundance of L. crispatus, while alpha diversity and L. iners were not significantly associated with sPTB. For each order of magnitude increase in the normalized relative abundance of L. crispatus, multivariable adjusted odds of sPTB decreased by approximately 20% (odds ratio, 0.81; 95% confidence interval, 0.70, 0.94). When we considered the impact of douching, associations between the microbiome and sPTB were limited to women who do not douche. We also observed strong intercorrelations between a range of maternal factors, including poverty, education, marital status, age, douching, and race, with microbiome effect sizes in the range of 1.8 to 5.2% in univariate models. Therefore, race may simply be a proxy for other socially driven factors that differentiate microbiome community structures. Future work will continue to refine reliable microbial biomarkers for preterm birth across diverse cohorts. IMPORTANCE Approximately 10% of all pregnancies in the United States end in preterm birth, and over 14% of pregnancies end in preterm birth among Black women. Knowledge on the associations between vaginal microbiome and preterm birth is important for understanding the potential cause and assessing risk of preterm birth. Our study is one of the largest studies performed to date to investigate the associations between vaginal microbiome and spontaneous preterm birth (sPTB), with stratified design for Black and White women. We found that the vaginal microbiome was different between Black and White women. The vaginal microbiome was associated with sPTB, and a lower abundance of L. crispatus increased the risk of sPTB independent of racial differences in microbial community structures. Furthermore, we also found that vaginal douching obscured the associations between vaginal microbiome, race, and preterm birth, suggesting that vaginal douching is an important factor to consider in future studies.

Project description:BackgroundDespite decades of attempts to link infectious agents to preterm birth, an exact causative microbe or community of microbes remains elusive. Nonculture 16S ribosomal RNA gene sequencing suggests important racial differences and pregnancy specific changes in the vaginal microbial communities. A recent study examining the association of the vaginal microbiome and preterm birth documented important findings but was performed in a predominantly white cohort. Given the important racial differences in bacterial communities within the vagina as well as persistent racial disparities in preterm birth, it is important to examine cohorts with varied demographic compositions.ObjectiveTo characterize vaginal microbial community characteristics in a large, predominantly African-American, longitudinal cohort of pregnant women and test whether particular vaginal microbial community characteristics are associated with the risk for subsequent preterm birth.Study designThis is a nested case-control study within a prospective cohort study of women with singleton pregnancies, not on supplemental progesterone, and without cervical cerclage in situ. Serial mid-vaginal swabs were obtained by speculum exam at their routine prenatal visits. Sequencing of the V1V3 region of the 16S rRNA gene was performed on the Roche 454 platform. Alpha diversity community characteristics including richness, Shannon diversity, and evenness as well as beta diversity metrics including Bray Curtis Dissimilarity and specific taxon abundance were compared longitudinally in women who delivered preterm to those who delivered at term.ResultsA total of 77 subjects contributed 149 vaginal swabs longitudinally across pregnancy. Participants were predominantly African-American (69%) and had a preterm birth rate of 31%. In subjects with subsequent term delivery, the vaginal microbiome demonstrated stable community richness and Shannon diversity, whereas subjects with subsequent preterm delivery had significantly decreased vaginal richness, diversity, and evenness during pregnancy (P < .01). This change occurred between the first and second trimesters. Within-subject comparisons across pregnancy showed that preterm birth is associated with increased vaginal microbiome instability compared to term birth. No distinct taxa were associated with preterm birth.ConclusionIn a predominantly African-American population, a significant decrease of vaginal microbial community richness and diversity is associated with preterm birth. The timing of this suppression appears early in pregnancy, between the first and second trimesters, suggesting that early gestation may be an ecologically important time for events that ordain subsequent term and preterm birth outcomes.

Project description:ObjectiveThis study aimed to identify immune states associated with a high risk of preterm birth by immunophenotyping in pregnant populations, and to elucidate the characteristics of immune subtypes and their relationships with preterm birth. Additionally, it sought to uncover the microbial composition and functional characteristics of immune states linked to preterm birth, and to evaluate the impact of bacterial interactions on the initiation of preterm birth.MethodsUtilizing 16S rRNA sequencing data and local immune factor expression data from a publicly available longitudinal pregnancy cohort, we conducted immunophenotyping through unsupervised clustering of the immune factors. We compared the differences in vaginal microbiota richness, diversity, and composition between identified immune subtypes using α and β diversity analysis. Signature microbiotas were identified using LEfSe analysis, and functional pathway enrichment variations were analyzed using PICRUSt2. Bidirectional mediation analysis was employed to construct a network of mediating roles, and preliminary in vitro validation of the Microbial-Cytokine-Preterm Birth pathway was performed to explore the effects of microbial and immune characteristics on vaginal epithelial cell function.ResultsPregnant women were categorized into three immune subtypes based on local immune status. Microbial functional analysis identified 31 distinct functional pathways, six of which were downregulated in the preterm birth and excessive inflammatory response group. Significant differences in vaginal microbial diversity and composition were observed among pregnant women with different immune subtypes. Bidirectional mediation analysis revealed multiple intermediary roles in preterm birth, highlighting C3b/iC3b and IL-8 in mid-pregnancy and IgE and IgM in late pregnancy.ConclusionThis study classified pregnant women into three immune subtypes, with the excessive inflammatory response subtype showing a higher predisposition to preterm birth. Mid-pregnancy immune status emerged as a key indicator of preterm birth risk, associated with the vaginal microbiome composition. Microorganisms affected the occurrence of preterm birth by modulating immune factor levels, with time-specific mediation roles observed. Lactobacillus crispatus demonstrated potential in protecting against preterm birth by modulating vaginal immune status.

Dataset Information

DeepMPTB: a vaginal microbiome-based deep neural network as artificial intelligence strategy for efficient preterm birth prediction.

Publications

DeepMPTB: a vaginal microbiome-based deep neural network as artificial intelligence strategy for efficient preterm birth prediction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets