Khan contains gene expression profiles of four types of small round blue cell tumours of childhood (SRBCT) published by Khan et al. (2001). It also contains further gene annotation retrieved from SOURCE at



Khan is dataset containing the following:

  • train:data.frame of 306 rows and 64 columns. The training dataset of 64 arrays and 306 gene expression values

  • test:data.frame, of 306 rows and 25 columns. The test dataset of 25 arrays and 306 genes expression values

  • gene.labels.imagesID:vector of 306 Image clone identifiers corresponding to the rownames of train and test.

  • train.classes:factor with 4 levels "EWS", "BL-NHL", "NB" and "RMS", which correspond to the four groups in the train dataset

  • test.classes:factor with 5 levels "EWS", "BL-NHL", "NB", "RMS" and "Norm" which correspond to the five groups in the test dataset

  • annotation:data.frame of 306 rows and 8 columns. This table contains further gene annotation retrieved from SOURCE in May 2004. For each of the 306 genes, it contains:

    • CloneIDImage Clone ID

    • UGClusterThe Unigene cluster to which the gene is assigned

    • SymbolThe HUGO gene symbol

    • LLIDThe locus ID

    • UGRepAccNucleotide sequence accession number

    • LLRepProtAccProtein sequence accession number

    • Chromosomechromosome location

    • Cytobandcytoband location


khan contains a filtered data of 2308 gene expression profiles as published and provided by Khan et al. (2001) on the supplementary web site to their publication OLD (site no longer found):

The data was copied from the made4 package (


Khan et al., 2001 used cDNA microarrays containing 6567 clones of which 3789 were known genes and 2778 were ESTs to study the expression of genes in of four types of small round blue cell tumours of childhood (SRBCT). These were neuroblastoma (NB), rhabdomyosarcoma (RMS), Burkitt lymphoma, a subset of non-Hodgkin lymphoma (BL), and the Ewing family of tumours (EWS). Gene expression profiles from both tumour biopsy and cell line samples were obtained and are contained in this dataset. The dataset downloaded from the website contained the filtered dataset of 2308 gene expression profiles as described by Khan et al., 2001. This dataset is available from the

In order to reduce the size of the MADE4 package, and produce small example datasets, the top 50 genes from the ends of 3 axes following bga were selected. This produced a reduced datasets of 306 genes.


Culhane AC, et al., 2002 Between-group analysis of microarray data. Bioinformatics. 18(12):1600-8.

Khan,J., Wei,J.S., Ringner,M., Saal,L.H., Ladanyi,M., Westermann,F., Berthold,F., Schwab,M., Antonescu,C.R., Peterson,C. et al. (2001) Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nat. Med., 7, 673-679.


#>                      Length Class      Mode     
#> train                 64    data.frame list     
#> test                  25    data.frame list     
#> train.classes         64    factor     numeric  
#> test.classes          25    factor     numeric  
#> annotation             8    data.frame list     
#> gene.labels.imagesID 306    -none-     character
#> cellType              64    -none-     character