Source:

@ARTICLE{Armstrong2002,

author = {Scott A Armstrong and Jane E Staunton and Lewis B Silverman and Rob Pieters and Monique L den Boer and Mark D Minden and Stephen E Sallan and Eric S Lander and Todd R Golub and Stanley J Korsmeyer},

title = {MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia.},

journal = {Nat Genet},

year = {2002},

volume = {30},

pages = {41--47},

number = {1},

month = {Jan},

url = {http://dx.doi.org/10.1038/ng765}

}

Original data: http://www.broad.mit.edu/mpr/publications/projects/Leukemia/expression_data.txt

Description:

The authors showed that clustering algorithms reveal that lymphoblastic leukemias with MLL translocations (MLL) can clearly be separated from conventional acute lymphoblastic (ALL) and acute myelogenous leukemias (AML).

From these, we formed two data sets with the following distribution of samples:

• Armstrong-V1: 24 ALL and 48 MLL
• Armstrong-V2: 24 ALL, 20 MLL and 28 AML

Parameters used in our filter:

• Armstrong-V1: l= 3 and c=30
• Armstrong-V2: l= 3 and c=18