Yeoh2002: Classification of acute lymphoblastic leukemia subtypes

In pbreheny/hdrm: High-Dimensional Regression Modeling

Classification of acute lymphoblastic leukemia subtypes

Description

There are six known categories of acute lymphoblastic leukemia (ALL). Unfortunately, the accurate assignment of patients to these subtypes is a difficult and expensive process, requiring intensive laboratory studies and the collective expertise of a number of professionals (usually only available at major medical centers).

In this study, bone marrow samples were obtained from pediatric patients, and gene expression measurements were taken. The goal is to determine ALL subtype from the gene expression data alone – if this can be done accurately, it would make it possible to diagnose ALL subtype at rural hospitals, in developing countries, etc.

The data has been preprocessed using vsn on the probe level and the probes have been summed up using the median polish technique.

Dimensions

• n = 248 observations

• p = 12,625 features

• K = 6 outcome categories

Annotation

• The object fData contains the associated gene names, gene symbols, and chromosome locations for the (mapped) probes in X. Rows of fData correspond to columns of X, and are named accordingly.

References

I obtained this data set from the stjudem package (on Bioconductor). The original reference is:

Yeoh et al.~(2002). Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell, 1: 133-143.

pbreheny/hdrm documentation built on Jan. 17, 2024, 8:53 p.m.