| 摘要 |
The invention relates to a method for preparing an optimised gene signature for assigning a NSCLC sample to one or more NSCLC classes, comprising subjecting a gene signature set forth in Table 12 herein to nearest shrunken centroid analysis to identify one or more subgroups of gene classifiers corresponding to one or more of classes 1 to 6 identified in Table 12, and validating the performance of the selected classifiers by K-fold or leave-one-out cross-validation.
|
