Diagnosis of multiple cancer types by shrunken centroids of gene expression

Proc Natl Acad Sci U S A. 2002 May 14;99(10):6567-72. doi: 10.1073/pnas.082099299.

Abstract

We have devised an approach to cancer class prediction from gene expression profiling, based on an enhancement of the simple nearest prototype (centroid) classifier. We shrink the prototypes and hence obtain a classifier that is often more accurate than competing methods. Our method of "nearest shrunken centroids" identifies subsets of genes that best characterize each class. The technique is general and can be used in many other classification problems. To demonstrate its effectiveness, we show that the method was highly efficient in finding genes for classifying small round blue cell tumors and leukemias.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, Non-P.H.S.
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Child
  • DNA, Neoplasm / analysis*
  • Discriminant Analysis
  • Gene Expression Profiling
  • Gene Expression*
  • Humans
  • Neoplasms / classification
  • Neoplasms / diagnosis*
  • Neoplasms / genetics
  • Precursor Cell Lymphoblastic Leukemia-Lymphoma / classification
  • Precursor Cell Lymphoblastic Leukemia-Lymphoma / diagnosis
  • Precursor Cell Lymphoblastic Leukemia-Lymphoma / genetics
  • Probability

Substances

  • DNA, Neoplasm