Metabolomics and Mass Spectrometry Studies

This study identifies age and sex-associated metabolite trajectories in longitudinal plasma samples from the WRAP cohort, revealing sex-specific changes in metabolites like sphingolipids and heritability patterns.

A short-term Western diet exposure in baboons caused long-term metabolic changes, suggesting a reset of the metabolic program.

An open-source cloud engine annotates imaging mass spectrometry data at controlled false discovery rate; publicly shared results reached over 3200 datasets and 6 million metabolite images by January 2019.

The paper provides efficient implementations of the Chemical Structural and Compositional Similarity metric in R and Python, significantly improving speed and usability for metabolomics analysis.

This study evaluates seven machine learning algorithms for predicting LC retention times across 36 metabolomics datasets, finding no single algorithm is optimal for all cases and highlighting the benefits of combining models.

PAIRUP-MS imputes and annotates unknown metabolite signals in mass spectrometry data, improving pathway analysis and meta-analysis power for metabolomics studies.

This study develops a targeted metabolomics approach using LC/MS/MS and LC/Q-TOF/MS to quantify purine metabolites in cultured cells, integrating automated label-free microscopy for normalization.

This paper introduces methods to improve metabolite annotation using substructure discovery and machine learning, along with a database for shared motifs.

This study improves NMR automatic metabolite profiling by using machine learning to predict signal parameters, reducing uncertainty and enhancing profiling quality.

netome provides tools for analyzing LC-MS metabolomics data, including quality control, peak alignment, and network-based statistical testing.

This study identifies metabolic and genetic signatures of obesity using metabolome analysis and whole genome sequencing in twin and volunteer cohorts, revealing clinically meaningful heterogeneity.

The paper improves metabolite identification in LC-MS datasets by incorporating in-source fragment information from MS/MS spectral libraries.

The paper integrates multimodal health data to identify novel biomarker signatures for health and disease, showing better predictive power for diabetes and hypertension than traditional biomarkers.

The paper evaluates missing data handling strategies in untargeted metabolomics data, finding that KNN imputation with variable pre-selection is robust across multiple evaluation schemes.

The paper identifies and addresses high peak density artifacts in Fourier transform mass spectra, which can mislead data analysis by creating false peaks.

The paper introduces a data-adaptive pipeline for filtering and normalizing metabolomics data, showing it outperforms traditional methods in reducing technical noise and preserving biological signals.

The paper develops a Bayesian graphical model to infer metabolite interactomes using molecular structure information and applies it to coronary artery disease data.

This paper introduces a framework for large-scale dynamic metabolome drug profiling in mammalian cells, demonstrating its application to study dichloroacetate's toxic effects in ovarian cancer cell lines.

The paper evaluates the use of Orbitrap All Ion Fragmentation (AIF) analysis with the MetaboList R package for metabolite profiling in leukemia and melanoma samples, showing comparable manual and automated results.

DIMEdb provides a unified web-based database for metabolite identification in direct infusion mass spectrometry, integrating over 80,000 metabolites from multiple sources.

This paper applies a Wisdom of Artificial Crowds (WoAC) strategy to feature selection in untargeted metabolomics for developing a blood-based diagnostic test for thrombotic myocardial infarction, achieving lower misclassification rates than Lasso and Elastic Net.

speaq 2.0 automates 1D NMR spectral processing and quantification, reducing user interaction and enabling reproducible metabolomics analysis.

GSimp is a Gibbs sampler-based method for imputing left-censored missing values in metabolomics data, showing superior performance in accuracy and statistical analysis.

The paper develops a quality control-based random forest regression method to correct signal drift in metabolomics and proteomics data, improving precision and statistical accuracy.

The paper compares imputation methods for metabolomics data with missing values and proposes a strategy with a web-tool.

R2DGC aligns and identifies peaks in 2D gas chromatography-mass spectrometry data without thresholds, using R.

Curatr is a web application that enables the manual curation of mass spectral libraries from LC-MS data, supporting export to standard formats and public access.

The paper develops methods to estimate false discovery rates in untargeted metabolomics annotations and shows how adjusting spectral matching parameters can significantly increase the number of annotations.

This study compares deep learning and machine learning methods for classifying estrogen receptor status in breast cancer metabolomics data, finding deep learning achieves higher accuracy.