Source:

@ARTICLE{Yeoh2002,

author = {Eng-Juh Yeoh and Mary E Ross and Sheila A Shurtleff and W. Kent Williams and Divyen Patel and Rami Mahfouz and Fred G Behm and Susana C Raimondi and Mary V Relling and Anami Patel and Cheng Cheng and Dario Campana and Dawn Wilkins and Xiaodong Zhou and Jinyan Li and Huiqing Liu and Ching-Hon Pui and William E Evans and Clayton Naeve and Limsoon Wong and James R Downing},

title = {Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling.},

journal = {Cancer Cell},

year = {2002},

volume = {1},

pages = {133--143},

number = {2},

month = {Mar},

}

Original data: http://www.stjuderesearch.org/data/ALL1/all_datafiles.html

Description:

Treatment of pediatric acute lymphoblastic leukemia (ALL) is based on the concept of tailoring the intensity of therapy to a patient's risk of relapse. To determine whether gene expression profiling could enhance risk assignment, the authors used oligonucleotide microarrays to analyze the pattern of genes expressed in leukemic blasts from 360 pediatric ALL patients. Distinct expression profiles identified each of the prognostically important leukemia subtypes, including T-ALL, E2A-PBX1, BCR-ABL, TEL-AML1, MLL rearrangement, and hyperdiploid >50 chromosomes. In addition, another ALL subgroup (OTHER) was identified based on its unique expression profile. In order to build our data set, we discard the samples belonging to OTHER. Also, we create another data set where we the samples are labeled either as T-ALL or B-ALL.

• Yeoh-V1: 43 T-ALL and B-ALL.
• Yeoh-V2 .. 43 T-ALL, 27 E2A-PBX1,15 BCR-ABL, 79 TEL-AML1, 20 MLL, and 64 hyperdiploid >50.

Parameters used in our filter:

• Yeoh-V1 and Yeoh-V2: l=3 and c=128