Summary of Study ST003587

This data is available at the NIH Common Fund's National Metabolomics Data Repository (NMDR) website, the Metabolomics Workbench, https://www.metabolomicsworkbench.org, where it has been assigned Project ID PR002217. The data can be accessed directly via it's Project DOI: 10.21228/M82821 This work is supported by NIH grant, U2C- DK119886.

See: https://www.metabolomicsworkbench.org/about/howtocite.php

This study contains a large results data set and is not available in the mwTab file. It is only available for download via FTP as data file(s) here.

Perform statistical analysis  |  Show all samples  |  Show named metabolites  |  Download named metabolite data  
Download mwTab file (text)   |  Download mwTab file(JSON)   |  Download data files (Contains raw data)
Study IDST003587
Study TitleComparison of Machine Learning Models for Metabolomic-Based Clinical Prediction of Preterm Birth
Study SummaryMachine learning (ML), with advancements in algorithms and computations, is seeing an increased presence in life science research. This study investigated several ML models' efficacy in predicting preterm birth using untargeted metabolomics from serum collected during the third trimester of gestation. Samples from 48 preterm and 102 term delivery mothers (1:2 ratio) from the All Our Families Cohort (Calgary, AB) were examined. Selected ML applications were used to examine the small-scale clinical dataset for both model performance and metabolite interpretation. Model performance was evaluated based on confusion matrices, receiver operating characteristic curves, and feature importance rankings. Conventional linear models, like Partial Least Squares Discriminant Analysis (PLS-DA) and linear logistic regression, showed moderate predictive potential with AUC-ROC around 0.60. Non-linear models, including Extreme Gradient Boosting (XGBoost) and artificial neural networks, had marginally improved predictive accuracy and strength. Resampling by bootstrapping was also examined. Among all MLs, bootstrap resampling enhanced XGBoost's performance the most, improving AUC-ROC (0.85, 95% CI:0.574-0.995, p<0.001) for the best fitted model. Feature importance analysis by Shapley Additive Explanations analysis consistently identified acylcarnitines and amino acid derivatives as significant metabolites. Findings underscored the complexity of modeling preterm birth prediction, suggesting a trial-and-error approach for model selection.
Institute
University of Calgary
Last NameHan
First NameYing Chieh
Address2500 University Drive NW
Emailyingchieh.han@ucalgary.ca
Phone17783848168
Submit Date2024-11-08
Num Groups2
Total Subjects150
Num Females150
Raw Data AvailableYes
Raw Data File Type(s)mzML
Analysis Type DetailLC-MS
Release Date2024-12-12
Release Version1
Ying Chieh Han Ying Chieh Han
https://dx.doi.org/10.21228/M82821
ftp://www.metabolomicsworkbench.org/Studies/ application/zip

Select appropriate tab below to view additional metadata details:

Project:

Project ID:PR002217
Project DOI:doi: 10.21228/M82821
Project Title:Metabolomic-Based Clinical Assessment of Preterm Birth
Project Summary:Machine learning (ML), with advancements in algorithms and computations, is seeing an increased presence in life science research. This study investigated several ML models' efficacy in predicting preterm birth using untargeted metabolomics from serum collected during the third trimester of gestation. Samples from 48 preterm and 102 term delivery mothers (1:2 ratio) from the All Our Families Cohort (Calgary, AB) were examined. Selected ML applications were used to examine the small-scale clinical dataset for both model performance and metabolite interpretation. Model performance was evaluated based on confusion matrices, receiver operating characteristic curves, and feature importance rankings. Conventional linear models, like Partial Least Squares Discriminant Analysis (PLS-DA) and linear logistic regression, showed moderate predictive potential with AUC-ROC around 0.60. Non-linear models, including Extreme Gradient Boosting (XGBoost) and artificial neural networks, had marginally improved predictive accuracy and strength. Resampling by bootstrapping was also examined. Among all MLs, bootstrap resampling enhanced XGBoost's performance the most, improving AUC-ROC (0.85, 95% CI:0.574-0.995, p<0.001) for the best fitted model. Feature importance analysis by Shapley Additive Explanations analysis consistently identified acylcarnitines and amino acid derivatives as significant metabolites. Findings underscored the complexity of modeling preterm birth prediction, suggesting a trial-and-error approach for model selection.
Institute:University of Calgary
Last Name:Han
First Name:Ying Chieh
Address:2500 University Drive NW, Calgary, Alberta, T2N 1N4, Canada
Email:yingchieh.han@ucalgary.ca
Phone:17783848168
Funding Source:NSERC

Subject:

Subject ID:SU003716
Subject Type:Human
Subject Species:Homo sapiens
Taxonomy ID:9606
Age Or Age Range:19-43
Weight Or Weight Range:44-116
Height Or Height Range:147-186
Gender:Female
Human Ethnicity:Caucasian
Species Group:Mammals

Factors:

Subject type: Human; Subject species: Homo sapiens (Factor headings shown in green)

mb_sample_id local_sample_id Termstatus Sample source
SA391403815139Preterm Birth Blood Serum
SA391404810506Preterm Birth Blood Serum
SA391405818540Preterm Birth Blood Serum
SA391406818520Preterm Birth Blood Serum
SA391407818409Preterm Birth Blood Serum
SA391408810526Preterm Birth Blood Serum
SA391409818368Preterm Birth Blood Serum
SA391410818323Preterm Birth Blood Serum
SA391411818237Preterm Birth Blood Serum
SA391412818224Preterm Birth Blood Serum
SA391413815179Preterm Birth Blood Serum
SA391414815161Preterm Birth Blood Serum
SA391415815076Preterm Birth Blood Serum
SA391416818577Preterm Birth Blood Serum
SA391417810585Preterm Birth Blood Serum
SA391418812580Preterm Birth Blood Serum
SA391419812566Preterm Birth Blood Serum
SA391420812523Preterm Birth Blood Serum
SA391421812517Preterm Birth Blood Serum
SA391422812478Preterm Birth Blood Serum
SA391423812471Preterm Birth Blood Serum
SA391424812462Preterm Birth Blood Serum
SA391425812459Preterm Birth Blood Serum
SA391426812285Preterm Birth Blood Serum
SA391427812359Preterm Birth Blood Serum
SA391428818575Preterm Birth Blood Serum
SA391429818274Preterm Birth Blood Serum
SA391430812342Preterm Birth Blood Serum
SA391431818822Preterm Birth Blood Serum
SA391432830909Preterm Birth Blood Serum
SA391433830872Preterm Birth Blood Serum
SA391434830850Preterm Birth Blood Serum
SA391435830762Preterm Birth Blood Serum
SA391436830687Preterm Birth Blood Serum
SA391437810387Preterm Birth Blood Serum
SA391438830640Preterm Birth Blood Serum
SA391439818614Preterm Birth Blood Serum
SA391440830560Preterm Birth Blood Serum
SA391441830505Preterm Birth Blood Serum
SA391442830408Preterm Birth Blood Serum
SA391443830390Preterm Birth Blood Serum
SA391444830635Preterm Birth Blood Serum
SA391445818716Preterm Birth Blood Serum
SA391446810453Preterm Birth Blood Serum
SA391447818626Preterm Birth Blood Serum
SA391448818781Preterm Birth Blood Serum
SA391449818684Preterm Birth Blood Serum
SA391450818732Preterm Birth Blood Serum
SA391451818036Term Birth Blood Serum
SA391452812455Term Birth Blood Serum
SA391453812435Term Birth Blood Serum
SA391454812431Term Birth Blood Serum
SA391455812412Term Birth Blood Serum
SA391456812404Term Birth Blood Serum
SA391457812379Term Birth Blood Serum
SA391458812397Term Birth Blood Serum
SA391459812373Term Birth Blood Serum
SA391460812369Term Birth Blood Serum
SA391461818556Term Birth Blood Serum
SA391462812352Term Birth Blood Serum
SA391463812458Term Birth Blood Serum
SA391464818706Term Birth Blood Serum
SA391465818695Term Birth Blood Serum
SA391466830757Term Birth Blood Serum
SA391467818010Term Birth Blood Serum
SA391468812479Term Birth Blood Serum
SA391469812483Term Birth Blood Serum
SA391470812498Term Birth Blood Serum
SA391471818371Term Birth Blood Serum
SA391472812550Term Birth Blood Serum
SA391473818758Term Birth Blood Serum
SA391474812350Term Birth Blood Serum
SA391475815111Term Birth Blood Serum
SA391476815124Term Birth Blood Serum
SA391477818197Term Birth Blood Serum
SA391478818789Term Birth Blood Serum
SA391479818593Term Birth Blood Serum
SA391480510369Term Birth Blood Serum
SA391481812347Term Birth Blood Serum
SA391482810499Term Birth Blood Serum
SA391483810473Term Birth Blood Serum
SA391484810482Term Birth Blood Serum
SA391485810489Term Birth Blood Serum
SA391486810492Term Birth Blood Serum
SA391487810495Term Birth Blood Serum
SA391488810497Term Birth Blood Serum
SA391489810502Term Birth Blood Serum
SA391490810467Term Birth Blood Serum
SA391491810503Term Birth Blood Serum
SA391492810509Term Birth Blood Serum
SA391493810513Term Birth Blood Serum
SA391494810514Term Birth Blood Serum
SA391495810523Term Birth Blood Serum
SA391496810537Term Birth Blood Serum
SA391497810472Term Birth Blood Serum
SA391498810466Term Birth Blood Serum
SA391499810542Term Birth Blood Serum
SA391500810388Term Birth Blood Serum
SA391501515122Term Birth Blood Serum
SA391502515123Term Birth Blood Serum
Showing page 1 of 2     Results:    1  2  Next     Showing results 1 to 100 of 150

Collection:

Collection ID:CO003709
Collection Summary:Serum samples were collected during the third trimester between 28 and 32 weeks' of gestation. The collected serum was centrifuged and stored at -80℃ storage until the day of sample assay.
Sample Type:Blood (serum)
Collection Location:Calgary
Storage Conditions:-80℃

Treatment:

Treatment ID:TR003725
Treatment Summary:The experimental group in this study were pregnant woman who later experienced preterm delivery. No additional treatment was implemented.

Sample Preparation:

Sampleprep ID:SP003723
Sampleprep Summary:Sample were prepared following the procedure described in the published manuscript titled “Maternal Acylcarnitine Disruption as a Potential Predictor of Preterm Birth in Primigravida” published in Nutrients (2024), 16(5), 595. doi: 10.3390/nu16050595.
Processing Storage Conditions:Room temperature
Extraction Method:Protein precipitation with methanol
Extract Storage:-80℃
Sample Resuspension:in 1:1 methanol:water
Sample Derivatization:no
Sample Spiking:no

Combined analysis:

Analysis ID AN005891
Analysis type MS
Chromatography type Reversed phase
Chromatography system Agilent QTOF 6545i
Column Waters ACQUITY UPLC HSS T3 (150 x 2.1 mm, 1.7 um)
MS Type ESI
MS instrument type QTOF
MS instrument name Agilent 6545 QTOF
Ion Mode POSITIVE
Units Peak Intensity

Chromatography:

Chromatography ID:CH004474
Chromatography Summary:Reverse Phase Positive ESI method
Instrument Name:Agilent QTOF 6545i
Column Name:Waters ACQUITY UPLC HSS T3 (150 x 2.1 mm, 1.7 um)
Column Temperature:40℃
Flow Gradient:Initiated with 5% B for 1.5 min, then a linear gradient of B from 5% to 100% for 14 min, followed by 100% B for 3 min. The gradient returned to the 5% B starting condition at the 17 min mark and equilibrated for 2 min to conclude the run
Flow Rate:0.4 mL/min
Solvent A:100% Water; 0.1% Formic acid
Solvent B:100% Acetonitrile; 0.1% Formic acid
Chromatography Type:Reversed phase

MS:

MS ID:MS005609
Analysis ID:AN005891
Instrument Name:Agilent 6545 QTOF
Instrument Type:QTOF
MS Type:ESI
MS Comments:MS spectrum was obtained in positive ionization mode betwenn 50 amd 1200 m/z. Peaks were labeled using XCMS web platform. Compound identifies were determined by inputting m/z for identified peaks in Human Metabolome Database for the most probable compound candidate based on tolerance threshold of 30 ppm.
Ion Mode:POSITIVE
  logo