Integrative Analysis of High Dimensional Gene Expression.pptVIP

May 24-25, 2004 NISS Metabolomics Workshop, 2005 Integrative Analysis of High Dimensional Gene Expression, Metabolite and Blood Chemistry Data Kwan R. Lee, Ph.D. and Lei A. Zhu, Amit Bhattacharyya, J. Alan Menius Biomedical Data Sciences GlaxoSmithKline kwan.lee@ Overview Systems Biology Challenges for Statisticians Possible solutions Example of integrative data analysis Summary and discussion Of mice and men Focusing on one platform may miss an obvious signal!!! How can efficacy failures be attacked? ‘Systems Biology’ approach to drug discovery Experimental Platforms Non-omics and Omics, what are they? Experimental Platforms Non-omics and Omics, what are they? (cont.) Traditional Blood Chemistry (non-omics) Gene Expression (transcriptomics) Metabolite (metabonomics) Lipid (lipomics) Protein (proteomics) Five Challenges Data Pre-processing High Dimensionality Multiple Testing for Marker Selection Data Integration Validation of the Prediction Model Challenge #1: Data Pre-processing Peak Alignment (NMR, LC/MS) Normalization (Gene Chip, NMR, LC/MS data) Why? Remove systematic bias in the data Normalization within the platform makes data comparable across samples Challenge # 2: High Dimensionality# of subjects # of variables Blood Chemistry: 9 markers Gene Expression: 22,000 probe sets Lipid LC/MS: 2, 000 peaks Metabolite LS/MS: 3,000 peaks NMR: 500 buckets Challenge #3: Multiple Testing in Variable Selection Challenge #4: Data integration Challenge #4: Data integration (cont.) Challenge #4: Data integration Example 1 Integration approach 1: Simple data integration Simply combining the platform data together, the platform with large amount of data and variability will dominate the other platforms PCA on Non-omics, Transcriptomics, and Combined. Non-omics (20) PCA on Non-omics, Transcriptomics, and Combined. Non-omics (20) Challenge #4: Data integrationExample 2 Integration approach 2: Integrate on selected markers 9 blood chemistry + 2000