The fact that the list of 519 rank-significant genes defined by the eight feature definer datasets was reduced to a pool of 283 genes could have led to the exclusion of some highly significant genes, due to the only reason that they were not shared by the AM, RO, TR, and MA datasets, thus impairing the possibility for development of a stronger model.
← all excerpts
Gene expression profiles for predicting metastasis in breast cancer: a cross-study comparison of classification methods.
1
—
—