Fig. 2From: A differential diagnosis between uterine leiomyoma and leiomyosarcoma using transcriptome analysisExpression patterns between datasets of 17 selected genes. The expression patterns of selected 17 genes in training set (a), validation set (b), 1st test set (c), and 2nd test set (d). Gene expression patterns in the normal (n = 114 samples from uterus, cervix, cervix uteri, endometrium) and leiomyoma groups (n = 17 samples) are stable, while they are dynamic in the leiomyosarcoma groups (n = 49 samples). The patterns were similar in all datasets. Color corresponds to the label. The y-axis represents GAPDH-scaled and zero-sum transformed expression values. These values were used to train the classifiersBack to article page