Three versions of the model are provided, each of ones differs by the gene expressions standardization method since it has an important impact on the subtype classification:
[object Object],[object Object],[object Object]
The model ssp2003.robust has been shown to reach the best concordance with the traditional clinical parameters (ER IHC, HER2 IHC/FISH and histological grade). However the use of this model is recommended only when the dataset is representative of a global population of breast cancer patients (no sampling bias, the 5 subtypes should be present).
References
T. Sorlie and R. Tibshirani and J. Parker and T. Hastie and J. S. Marron and A. Nobel and S. Deng and H. Johnsen and R. Pesich and S. Geister and J. Demeter and C. Perou and P. E. Lonning and P. O. Brown and A. L. Borresen-Dale and D. Botstein (2003) "Repeated Observation of Breast Tumor Subtypes in Independent Gene Expression Data Sets", Proceedings of the National Academy of Sciences, 1(14):8418--8423