| Abstract Title: | MsCompare: An Untargeted GC/MS Metabolomics Platform for Quality Control, Precise Deconvolution and Data Analysis |
| Abstract Type: | Poster |
| Session Choice: | Big Data Chemometrics and Method Development(In-Silico)(KVCV) |
| Presenter Name: | Dr Marco Ruijken |
| Company/Organisation: | MsMetrix |
| Country: | Netherlands |
Abstract Information :
A GC/MS workflow for Metabolomics includes a number of distinct steps:
Experimental Design, Sampling, Sample Preparation, Data Analysis, Identification and Data Interpretation. The MsCompare platform includes all tools to properly control each step in this workflow.
One of the key issues in this field is the precise and sensitive detection of all components present in a series of samples. GC/MS deconvolution remains by far the most difficult step for low level components, especially when highly similar co-eluting or nearly co-eluting compounds are present. In these cases, precise GC/MS Peak Detection and Deconvolution is necessary with minimal user interference. Proper deconvolution also allows for correct identification of all components.
Another problematic area in GC/MS Metabolomics studies might be the proper alignment of chromatograms before the actual data processing starts. Depending on the application, we often see individual components in GC/MS having bad peaks shapes or bad reproducibility regarding retention times. MsCompare contains a number of alignment algorithms to correct for this behavior.
Data analysis in MsCompare comprises both Univariate and Multivariate analysis methods like PCA, PLS-DA, Clustering etc. However, it will be shown that for many cases, due to the high selectivity of GC/MS, univariate analysis methods are adequate in solving the main questions.
Examples from a number of different studies (small and large) will be given, showing an overview of the workflow and implemented tools.
