Systematic Analysis of DNA Microarray Data: Ordering and Interpreting Patterns of Gene Expression
- Paul J. Planet1,
- Rob DeSalle2,
- Mark Siddall2,
- Timothy Bael3,
- Indra Neil Sarkar4, and
- Scott E. Stanley5,6
- 1 Department of Microbiology, Columbia University College of Physicians and Surgeons, New York, New York 10032, USA; 2Division of Invertebrate Zoology, American Museum of Natural History, New York, New York 10024, USA; 3 Department of Internal Medicine, Columbia-Presbyterian Medical Center, New York, New York 10032, USA; 4 Department of Medical Informatics, College of Physicians and Surgeons, Columbia University, New York, New York 10032, USA; 5 Genaissance Pharmaceuticals, New Haven, Connecticut 06511, USA
This extract was created in the absence of an abstract.
From Gene Expression to Trees: The View From Systematics
Systematic methods have contributed greatly to the fields of comparative and evolutionary biology as tools for finding natural patterns in molecular and morphological data. Computational methods in systematic biology are designed to order data in terms of a hierarchy of relationships from which evolutionary history can be inferred (Miyamoto and Cracraft 1991). Gene expression profiles present a type of data that seems analogous to other types of data used in systematic analyses. Many recent attempts to organize the impressive amount of information in microarray studies have relied on techniques borrowed from systematic biology that offer quick computation and organization of the data (Eisen et al. 1998; Eisen and Brown 1999; Bassett et al. 1999). However, the goals of gene expression analyses differ significantly from the goals of reconstructing evolutionary history. Here, we review the systematic techniques currently applied to large gene expression data sets and discuss the ramifications of applying these and other systematic techniques to expression profiles. We suggest that the large body of work generated by phylogenetic systematists over the last few decades is relevant to understanding which techniques might be best applied to attain the specific goals of gene expression analysis. We believe these techniques could form a practical and theoretical framework for assessing the outcomes of DNA microarray studies.
The major goal of systematic analysis is to extract order from data that gives clues about biological reality. In phylogenetic systematics, the biological reality is the evolutionary relationships among taxa, and the biological order is the hierarchical pattern in the data that tracks the lineage splitting and divergence represented by a dendrogram or tree. In contrast, systematic treatment of microarray data assumes that order intrinsic to gene expression profiles will yield insights into molecular, cellular, and tissue level processes and functions. This approach, in turn, …
