Metabolomics Metabolomics Metabolome Reflects the State of the Cell, Organ or Metabolome Reflects the State of the Cell, Organ or Organism Organism Change in the metabolome is a direct consequence of protein activity changes • Not necessarily true for genomic, proteomic or transcriptomic changes Disease, environmental factors, Drugs, etc., perturbs the state of the metabolome
22
Embed
Metabolomics Metabolome Reflects the State of the Cell, Organ or Organism Change in the metabolome is a direct consequence of protein activity changes.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
MetabolomicsMetabolomicsMetabolome Reflects the State of the Cell, Organ or OrganismMetabolome Reflects the State of the Cell, Organ or Organism
Change in the metabolome is a direct consequence of protein activity changes • Not necessarily true for genomic, proteomic or transcriptomic changes
Disease, environmental factors, Drugs, etc., perturbs the state of the metabolome • Provides a system-wide view of the organism or cell’s response
Differential NMR MetabolomicsDifferential NMR MetabolomicsThe Role of NMR Signal-to-Noise in PCA Clustering
Increasing Number of NMR Scans Increasing Number of NMR Scans (S/N)(S/N)
Differential NMR MetabolomicsDifferential NMR MetabolomicsHow to Quantify the Statistical Significance of Cluster Separations?
Analyze Metabolomic Data Using Tree Diagrams• Calculate distances between cluster centers distance matrix
Apply Standard Boot-Strapping Methods• Randomize selection of cluster members to determine cluster center• Generate 100 different distance matrices 100 different trees consensus tree• Bootstrap number -> how many times the consensus node appears in the set of 100 trees
Differential NMR MetabolomicsDifferential NMR MetabolomicsBootstrap Number and Statistical Significance of Cluster Separations
Larger the Distance Between Clusters More Significant
• Larger bootstrap or smaller p-value• > 50% is significant
More Data Points Easier to Distinguish Between Clusters
• more data points (solid line)
Sample Replicates Affects Class DistinctionSample Replicates Affects Class Distinction
Increasing number of replicates
6
810
Significant increase in statistical significance of cluster from a modest increase in number or replicates
Ellipses and Tree Diagrams Define ClassesEllipses and Tree Diagrams Define Classes
P-value on each node identifies statistical significance (< 0.001) of cluster Ellipses represent 95% confidence limits from a normal distribution
Orthogonal partial least squares discriminant analysis (OPLS-DA)• a non-linear variant of PCA that minimizes class (group) variations• S-plots and loadings identify which “bins” (NMR chemical shifts – metabolites) are strongly correlated with class separation
S-plotsS-plotsloadingsloadings
Metabolite Identification
Overlay of 2D 1H-13C HSQC spectra for wild-type (red) and aconitase mutant (black)
Grow cells in the presence of a 13C-labeled metabolite
Only observe metabolites derived from the 13C-labeled
metabolite provided to the cells
O
OH
OH
OH
OH
HO
NH2
O
OH
Hu et al. (2011) J. Am. Chem. Soc. 133:1662-1665
Convert Peak Intensities to Convert Peak Intensities to Concentrations (HSQC0)Concentrations (HSQC0)
Our 2D 1H-13C HSQC calibration curve
Convert Peak Intensities to Convert Peak Intensities to Concentrations (HSQC0)Concentrations (HSQC0)
Can now compare changes between metabolites
Convert Concentrations to HeatmapConvert Concentrations to Heatmap
Provides two-levels of hierarchal clustering• Identifies replicates with same overall changes• Identifies metabolites with correlated changes between replicates
Provides a simple view of a large amount of dataCalculated with a statistical package, like R