Udskriv Udskriv
Switch language
Region Hovedstaden - en del af Københavns Universitetshospital

Data Pipeline for Clinical Diabetes Metabolomics - From Targeted Mass-Spectra to Patient Stratification

Publikation: KonferencebidragKonferenceabstrakt til konferenceForskningpeer review

  1. Effect of liraglutide on expression of inflammatory genes in type 2 diabetes

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  2. Ceramides and phospholipids are downregulated with liraglutide treatment: results from the LiraFlame randomized controlled trial

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

  3. Effects of Dapagliflozin in Stage 4 Chronic Kidney Disease

    Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Vis graf over relationer
Introduction Targeted metabolomics is entering diabetes clinics to provide a blood-molecular snapshot reflecting patient’s diet, lifestyle, disease status and treatment. High-throughput quantitative analysis of relevant metabolites can also provide new insights into disease subtypes and comorbidities that are otherwise hard to characterize. Methods We quantify metabolites using a selected reaction monitoring method built on ultra-high performance liquid chromatography triple-quadrupole mass spectrometry. We have developed a flexible and methods-rich computational pipeline for pre-processing, normalization, calibration, quality control and statistical analysis of the mass spectral data--all coupled with dynamic reproducible reporting. The pipeline is built on Skyline, which is a cutting-edge targeted mass spectrometry tool broadly used in proteomics, and R, which as a flexible and reproducible statistical computing platform. We demonstrate the approach with results from a small cross-sectional study of type 1 diabetes (T1D) patients with or without a kidney complication (macro- or normo-albuminuria; N=25+25, respectively, all with 3 replicate analyses). We test for metabolite-wise differences with a random-effects model, taking into account sample replication and adjusting for 14 relevant clinical variables. Results We produced an automated quality control report with visualizations of the measured metabolites individually and as whole. The report covers peak picking, technical variation, normalization and calibration. In the study, 7 amino acids were elevated in macroalbuminuric patients (FDR<0.05), and also two bile acids and kynurenine were deregulated. Conclusion We have developed a high-throughput pre-processing and quality control pipeline for targeted metabolomics. These solutions are contributing to the implementation of precision medicine at a diabetes clinic.
Publikationsdato28 jun. 2018
StatusUdgivet - 28 jun. 2018
BegivenhedMetabolomics 2018: The 14th International Conference of the Metabolomics Society - Washington State Convention Center, Seattle, USA
Varighed: 24 jun. 201828 jun. 2018


KonferenceMetabolomics 2018
LokationWashington State Convention Center


Metabolomics 2018: The 14th International Conference of the Metabolomics Society


Seattle, USA

Begivenhed: Konference

ID: 54966318