Jan Stanstrup
Building reproducible R tools and pipelines for LC-MS metabolomics
I am a metabolomics bioinformatician with 15 years of experience in LC-MS data preprocessing, data analysis of untargeted metabolomics data and compound annotation.
I build and maintain R packages and complex Shiny apps for the metabolomics community that span XCMS-based preprocessing workflows and retention time prediction to automated quality control.
Selected Projects
QC4Metabolomics
Automated quality control for LC-MS metabolomics experiments in real-time — catches drift and instrumental problems early.
PredRet
Retention time prediction system that maps RTs between LC-MS chromatographic systems, enabling portable compound libraries and cross-system mapping.
xcmsVis
Modern, interactive visualization for XCMS metabolomics data using ggplot2 and Plotly.
See the Code page for the full list of software.
Current Work
I work in the Microbiome & Metabolomics group led by Henrik M. Roager at NEXS, University of Copenhagen, where we study how individuals differ in their metabolic and microbial responses to diet — combining metabolomics, gut microbiome profiling, and human dietary intervention studies. My main research topics are:
Quality control & reproducibility
Automated real-time QC of LC-MS experiments. Here I built an automated system (QC4Metabolomics) that picks up new samples as they have been analyzed instrumentally and it then tracks performance metrics that are presented in a Shiny web-interface.
Active research includes adding support for more data types and quality metrics.
Retention time prediction
Cross-system RT projection so compound libraries can be shared between LC-MS instruments.
Active research includes a new version of PredRet that better handles the increased number of chromatographic systems the database now includes. The plan is to separate the model calculations from the database to simplify the setup and use RepoRT as the base database.
Compound annotation
MS/MS spectral matching pipelines with confidence reporting for LC-MS features.
Active development includes XCMS Annotator, which aims to give a more user-friendly way to use the R-based infrastructure to annotate untargeted datasets. MScurate is a tool to help supervise spectral library building in an efficient way.
Nutritional metabolomics
Personalized dietary responses through metabolomics and intervention studies.
I help establish the appropriate pre-processing pipelines and data analysis strategies for our studies. Recent development work includes using XCMS to process targeted analyses of short-chain fatty acids and a targeted list of biomarkers of food intake.
Get in touch
I’m always interested in talking about metabolomics, retention time prediction, R tooling, and reproducible pipelines.
For academic collaborations — joint projects, co-supervision, or method development — please write to jst@nexs.ku.dk.
For everything else — informal questions, prototype ideas, workshop requests — I’m happy to chat at stanstrup@gmail.com. Don’t worry about whether your inquiry “fits”; just write.