Figure 1
Figure 1. Expression-based classification of FLT3-ITD status in NK-AML training and test sets. Heatmap representations of gene-expression levels of the PAM-derived 20-gene FLT3-ITD predictor, shown for the NK-AML (A) training and (B) independent test sets. Mean-centered log2 gene-expression ratios are depicted by ratio color scale, indicated. Gray denotes poorly measured data. Genes are ordered by fractional contribution to the PAM predictor, and samples by PAM probabilities of FLT3-ITD classification, also displayed graphically above. Molecular alterations are coded by shaded boxes: FLT3-ITD (■), not FLT3-ITD (; FLT3-ITD length > 48 nt (median; ■); ≤ 48 nt (); not FLT3-ITD (); FLT3-ITD/WT ratio > 0.587 (median; ■), FLT3-ITD/WT ratio ≤ 0.587 (); FLT3-TKD (■), not FLT3-TKD (); not available (). Misclassified samples are indicated by gray arrowhead (misclassified as not FLT3-ITD) or black arrowhead (misclassified as FLT3-ITD).

Expression-based classification of FLT3-ITD status in NK-AML training and test sets. Heatmap representations of gene-expression levels of the PAM-derived 20-gene FLT3-ITD predictor, shown for the NK-AML (A) training and (B) independent test sets. Mean-centered log2 gene-expression ratios are depicted by ratio color scale, indicated. Gray denotes poorly measured data. Genes are ordered by fractional contribution to the PAM predictor, and samples by PAM probabilities of FLT3-ITD classification, also displayed graphically above. Molecular alterations are coded by shaded boxes: FLT3-ITD (■), not FLT3-ITD (; FLT3-ITD length > 48 nt (median; ■); ≤ 48 nt (); not FLT3-ITD (); FLT3-ITD/WT ratio > 0.587 (median; ■), FLT3-ITD/WT ratio ≤ 0.587 (); FLT3-TKD (■), not FLT3-TKD (); not available (). Misclassified samples are indicated by gray arrowhead (misclassified as not FLT3-ITD) or black arrowhead (misclassified as FLT3-ITD).

Close Modal

or Create an Account

Close Modal
Close Modal