NAME: Leukemia data TYPE: Observational study SIZE: 72 observations, 21 variables DESCRIPTIVE ABSTRACT: Modified version of the data discussed in Golub et al (1999). The data has been preprocessed and the 20 most important genes as determined by the BW criterion are included (Alexandritis et al, 2004). SOURCES: Golub TR, Slonin DK, et al. (1999). Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression Monitoring. Science, 286: 531-537 Alexandritis R, Lin S, and Irwin ME (2004). Class Discovery and Classification of Tumor Samples Using Mixture Modeling of Gene Expression Data. Bioinformatics, in press. VARIABLE DESCRIPTIONS: 1-20. Processed gene expression for genes 1 to 20. 21. Truth form of leukemia