Downloaded 116 times
Uploaded byDmitry Grapov
0 introduction
This document introduces a series of tutorials for metabolomic data analysis. It discusses important goals like hypothesis generation, data acquisition, processing, exploration, classification and prediction. It covers topics like univariate vs multivariate analysis, data quality metrics, clustering, principal component analysis, partial least squares modeling, and biological interpretation through metabolite enrichment and network mapping. The overall document provides a high-level overview of the key concepts and analytical approaches that will be covered in more detail in the tutorial series.
0 introduction
- 1. Introduction Introduction to Metabolomic DataAnalysis Dmitry Grapov, PhD
- 2. Introduction Important •This is anintroduction to a series of 8 tutorials for metabolomic data analysis •Download all the required files and software here: https://sourceforge.net/projects/teachingdemos/files/Winter%202014%20LC-MS%20and%20Statistics%20Course/ •Then follow the directions in the software/startup.R to launch all accompanying software
- 3.
- 4. Analysis at theMetabolomic Scale
- 5. Cycle of ScientificDiscovery Hypothesis Hypothesis Generation Data Acquisition Data Processing Data Analysis Data
- 6. Univariate vs. Multivariate Multivariate PredictiveModeling Group 2 Group 1 Univariate Hypothesis testing (t-Test, ANOVA, etc.) PCA O-/PLS/-DA
- 7.
- 8. Data Analysis Goals Exploration Classification •Are there any trends in my data? – analytical sources – meta data/covariates • Useful Methods – matrix decomposition (PCA, ICA, NMF) – cluster analysis • Differences/similarities between groups? – discrimination, classification, significant changes • Useful Methods – analysis of variance (ANOVA), mixed effects models – partial least squares discriminant analysis (O-/PLS-DA) – Others: random forest, CART, SVM, ANN • What is related or predictive of my variable(s) of interest? – Regression, correlation • Useful Methods – correlation – partial least squares (O-/PLS) Prediction
- 9. Data Complexity Meta Data m n variables Experimental Design = complexity samples Data m-D 1-D2-D Variable # = dimensionality
- 10. Univariate Qualities •length (samplesize) •center (mean, median, geometric mean) •dispersion (variance, standard deviation) •range (min / max), •quantiles •shape (skewness, kurtosis, normality, etc.) standard deviation mean
- 11. Data Quality Metrics • Precision •Accuracy Remedies • normalization • outliers detection *Start lab 1-statistical analysis
- 12. Univariate Analyses •Identify differencesin sample population means •sensitive to distribution shape •parametric = assumes normality •error in Y, not in X (Y = mX + error) wide •optimal for long data •assumed independence •false discovery rate (FDR) long n-of-one
- 13. False Discovery Rate(FDR) Type I Error: False Positives •Type II Error: False Negatives •Type I risk = •1-(1-p.value)m m = number of variables tested FDR correction • p-value adjustment or estimate of FDR (Fdr, q-value) Bioinformatics (2008) 24 (12):1461-1462
- 14. Achieving “significance” isa function of: significance level (α) and power (1-β ) effect size (standardized difference in means) sample size (n) *finish lab 1-statistical analysis
- 15. Clustering Identify •patterns •group structure •relationships •Evaluate/refine hypothesis •Reducecomplexity Artist: Chuck Close
- 16. Cluster Analysis Use theconcept similarity/dissimilarity to group a collection of samples or variables Linkage Approaches •hierarchical (HCA) •non-hierarchical (k-NN, k-means) •distribution (mixtures models) •density (DBSCAN) •self organizing maps (SOM) Distribution k-means Density
- 17. Hierarchical Cluster Analysis •similarity/dissimilarity defines “nearness” or distance euclidean manhattan Mahalanobis non-euclidean X X X * Y Y Y
- 18. Hierarchical Cluster Analysis Agglomerative/linkagealgorithm defines how points are grouped single complete centroid average
- 19.
- 20. Hierarchical Cluster Analysis Howdoes my metadata match my data structure? Exploration *finish lab 2-Cluster Analysis Confirmation
- 21. Projection of Data Thealgorithm defines the position of the light source Principal Components Analysis (PCA) • unsupervised • maximize variance (X) Partial Least Squares Projection to Latent Structures (PLS) • supervised • maximize covariance (Y ~ X) James X. Li, 2009, VisuMap Tech.
- 22. Interpreting PCA Results Varianceexplained (eigenvalues) Row (sample) scores and column (variable) loadings
- 23. How are scoresand loadings related?
- 24. Centering and Scaling PMID:16762068 *finish lab 3-Principal Components Analysis
- 25. Use PLS totest a hypothesis Partial Least Squares (PLS) is used to identify planes of maximum correlation between X measurements and Y (hypothesis) PLS PCA time = 0 120 min.
- 26. Modeling multifactorial relationships ~two-way ANOVA dynamicchanges among groups
- 27. PLS Related Objects Model •dimensions,latent variables (LV) •performance metrics (Q2, RMSEP, etc) •validation (training/testing, permutation, cross-validation) •orthogonal correction Samples •scores •predicted values •residuals Variables •Loadings •Coefficients, summary of loadings based on all LVs •VIP, variable importance in projection •Feature selection
- 28. “goodness” of themodel is all about the perspective Determine in-sample (Q2) and outof-sample error (RMSEP) and compare to a random model •permutation tests •training/testing *finish lab 4-Partial Least Squares and lab 5-Data Analysis Case Study
- 29. Biological Interpretation Projection ormapping of analysis results into a biological context. • Visualization • Enrichment • Networks – biochemical – structural – spectral – empirical
- 30. Identification of alterationsin biochemical domains Organism specific biochemical relationships and information Multiple organism DBs •KEGG •BioCyc •Reactome •Human •HMDB •SMPDB *finish lab 6-Metabolite Enrichment Analysis
- 31. Network Mapping 1. Generate Connections 2.Calculate Mappings 3. Create Network Grapov D., Fiehn O., Multivariate and network tools for analysis and visualization of metabolomic data, ASMS, June 08, 2013, Minneapolis, MN
- 32. Connections and Contexts Biochemical (substrate/product) •Databaselookup •Web query Chemical (structural or spectral similarity ) •fingerprint generation BMC Bioinformatics 2012, 13:99 doi:10.1186/1471-2105-13-99 Empirical (dependency) •correlation, partial-correlation
- 33. Mapping Analysis Results Analysisresults Network Annotation *finish lab 7-Network Mapping I Mapped Network
- 34.
- 35.
- 36. Mass Spectral Connections WatrousJ et al. PNAS 2012;109:E1743-E1752 *finish lab 8-Network Mapping II