Fig 2.

Exploratory margin tree analysis. Margin tree classification is a supervised multiclass support vector machine classification method.¹⁸ The margin tree program was applied to the stage I data set of 2,096 samples, characterized by their 18 class subtype labels (C1 to C18), and was based on 54,630 probe sets. B-ALL, B-cell acute lymphoblastic leukemia; MLL, myeloid/lymphoid or mixed-lineage leukemia; pre, precursor; c-ALL, childhood acute lymphoblastic leukemia; T-ALL, T-cell acute lymphoblastic leukemia; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; kt., karyotype; abn., abnormality; CLL, chronic lymphocytic leukemia; CML, chronic myelogenous leukemia; MDS, myelodysplastic syndrome.