Microarray studies

A microarray experiment captures the expression of thousands of genes simultaneously. When interested in changes over time of expression levels of many genes in a cell, one can use several microarrays to obtain a large number of gene expression profiles over time simultaneously. My primary focus in the analysis of multiple curves is to cluster the expression profiles based on similar behavior over time or cluster by shape. In the microarray setting, genes with similar expression profiles are coexpressed. Co-expression can suggest functional pathways and interactions between genes. Other research in this area is the discovery of a cross-hybridization artifact intrinsic to spotted single-dye cDNA as a result of cDNA containing 5'-end sequences of consecutive thymidine (dT) residues.

Selected Papers:
1. Handley, D, Serban, N., Peters, D., O'Doherty, R., Feild, M., Wasserman, L., Spirites, P., Scheines, R., Glymour, C. (2005), "Evidence of Cross-Hybridization Artifact in Expressed Sequence Tags (ESTs) on cDNA microarrays", Genomics, 83(6):1169-75. [.pdf ]
2. Serban, N., Wasserman, L. (2005), "CATS: Cluster Analysis by Transformation and Smoothing", , Journal of the American Statistical Association, 100. [.pdf ]