Human oral metagenome
Ontology highlight
ABSTRACT:
PROVIDER: PRJNA985119 | ENA |
REPOSITORIES: ENA
Similar Datasets
Project description:Spontaneous preterm birth (sPTB) is a leading cause of maternal and neonatal morbidity and mortality, yet its prevention and early risk stratification are limited. Previous investigations have suggested that vaginal microbes and metabolites may be implicated in sPTB. Here we performed untargeted metabolomics on 232 second-trimester vaginal samples, 80 from pregnancies ending preterm. We find multiple associations between vaginal metabolites and subsequent preterm birth, and propose that several of these metabolites, including diethanolamine and ethyl glucoside, are exogenous. We observe associations between the metabolome and microbiome profiles previously obtained using 16S ribosomal RNA amplicon sequencing, including correlations between bacteria considered suboptimal, such as Gardnerella vaginalis, and metabolites enriched in term pregnancies, such as tyramine. We investigate these associations using metabolic models. We use machine learning models to predict sPTB risk from metabolite levels, weeks to months before birth, with good accuracy (area under receiver operating characteristic curve of 0.78). These models, which we validate using two external cohorts, are more accurate than microbiome-based and maternal covariates-based models (area under receiver operating characteristic curve of 0.55-0.59). Our results demonstrate the potential of vaginal metabolites as early biomarkers of sPTB and highlight exogenous exposures as potential risk factors for prematurity.
2022-11-21 | MTBLS702 | MetaboLights
Project description:Oral tongue squamous cell carcinomas (OTSCC) are a homogenous group of aggressive tumors in the head and neck region, with a rising incidence among younger population. The role of altered DNA methylation in OTSCC and its link with clinical parameters has not been fully assessed yet. We performed genome-wide methylation analysis of oral tongue primary tumors (n = 52) using 485, 512 probes and correlated altered methylation with differences in gene expression. We used an ensemble machine-learning algorithm to identify differentially methylated probes and regions predictive of survival, risk habits, nodal status, tumor stage, and HPV infection followed by validation using data from the cancer genome atlas (TCGA) project on oral tongue (n = 24) and tumors from all subsites of head and neck region (n = 50). Bisulphite converted DNA from the 52 tumor:matched control sample pairs were hybridised to the Illumina Infinium 450k Human Methylation Beadchip
2016-06-01 | E-GEOD-75537 | biostudies-arrayexpress
Project description:Oral tongue squamous cell carcinomas (OTSCC) are a homogenous group of aggressive tumors in the head and neck region, with a rising incidence among younger population. The role of altered DNA methylation in OTSCC and its link with clinical parameters has not been fully assessed yet. We performed genome-wide methylation analysis of oral tongue primary tumors (n = 52) using 485, 512 probes and correlated altered methylation with differences in gene expression. We used an ensemble machine-learning algorithm to identify differentially methylated probes and regions predictive of survival, risk habits, nodal status, tumor stage, and HPV infection followed by validation using data from the cancer genome atlas (TCGA) project on oral tongue (n = 24) and tumors from all subsites of head and neck region (n = 50).
2016-06-01 | GSE75537 | GEO
Project description:Oral tongue squamous cell carcinomas (OTSCC) are a homogenous group of aggressive tumors in the head and neck region, with a rising incidence among younger population. The role of altered DNA methylation in OTSCC and its link with clinical parameters has not been fully assessed yet. We performed genome-wide methylation analysis of oral tongue primary tumors (n = 52) using 485, 512 probes and correlated altered methylation with differences in gene expression. We used an ensemble machine-learning algorithm to identify differentially methylated probes and regions predictive of survival, risk habits, nodal status, tumor stage, and HPV infection followed by validation using data from the cancer genome atlas (TCGA) project on oral tongue (n = 24) and tumors from all subsites of head and neck region (n = 50).
2016-06-01 | GSE75539 | GEO
Project description:Oral tongue squamous cell carcinomas (OTSCC) are a homogenous group of aggressive tumors in the head and neck region, with a rising incidence among younger population. The role of altered DNA methylation in OTSCC and its link with clinical parameters has not been fully assessed yet. We performed genome-wide methylation analysis of oral tongue primary tumors (n = 52) using 485, 512 probes and correlated altered methylation with differences in gene expression. We used an ensemble machine-learning algorithm to identify differentially methylated probes and regions predictive of survival, risk habits, nodal status, tumor stage, and HPV infection followed by validation using data from the cancer genome atlas (TCGA) project on oral tongue (n = 24) and tumors from all subsites of head and neck region (n = 50).
2016-06-01 | GSE75538 | GEO
Project description:The objectives of this study were to establish a microbiome profile for oral epithelial dysplasia using archival lesion swab samples to characterize the community variations and the functional potential of the microbiome using 16S rRNA gene sequencing
2022-09-30 | GSE198811 | GEO
Project description:HobPre predicts the oral bioavailability of small molecules in humans. It has been trained using public data on ~1200 molecules (Falcón-Cano et al, 2020, complemented with other literature and ChEMBL compounds).
Model Type: Predictive machine learning model.
Model Relevance: Predicts Probability of a compound having high oral bioavailability.
Model Encoded by: Hellen Namulinda (Ersilia)
Metadata Submitted in BioModels by: Zainab Ashimiyu-Abdusalam
Implementation of this model code by Ersilia is available here:
https://github.com/ersilia-os/eos2lqb
2024-06-03 | MODEL2406030001 | BioModels
Project description:The objective of this study was to develop and validate predictive models to assess the risk of death in patients with acute diquat (DQ) poisoning using innovative machine learning techniques. The study analyzed data from 201 consecutive patients admitted for deliberate oral intake of DQ between February 2018 and August 2023 at the First Hospital and Shengjing Hospital of China Medical University. Initial clinical data were collected, and four machine learning methods—logistic regression, random forest, support vector machine (SVM), and gradient boosting—were applied to build the prediction models. The dataset was split into a training set and a test set in an 8:2 ratio. The performance of these models was evaluated in terms of discrimination, calibration, and clinical decision curve analysis (DCA). Additionally, the SHapley Additive ExPlanations (SHAP) interpretation tool was used to provide an intuitive explanation of the risk of death in patients with DQ poisoning by calculating the contribution and impact of each feature on the final prediction. The areas under the receiver operating characteristic curves (AUCs) for the models were 0.91 for logistic regression, 0.98 for random forest, 0.96 for SVM, and 0.94 for gradient boosting. The random forest model demonstrated the best predictive performance with the highest AUC of 0.98, highest F1-score (0.90), highest Matthews correlation coefficient (0.79), highest accuracy (0.90), and the lowest Brier score (0.07). The study concludes that these machine learning models, combined with SHAP, provide reliable and interpretable tools for predicting the death risk in patients with acute DQ poisoning, with the random forest model being identified as the best performing model.
2024-07-18 | MODEL2407180001 | BioModels