Single-cell developmental classification of B cell precursor acute lymphoblastic leukemia at diagnosis reveals predictors of relapse

This dataset supports the publication "Single-Cell Developmental
Classification of B-Cell Precursor Acute Lymphoblastic Leukemia at
Diagnosis Reveals Predictors of Relapse" (Nature Medicine, 2018). It
contains single-cell mass cytometry measurements from 60 pediatric
patients diagnosed with B-cell precursor acute lymphoblastic leukemia
(BCP-ALL), along with data from 5 healthy bone marrow donors. Using a
panel of 35 surface and intracellular proteins, each leukemia cell was
classified into its closest developmental B-cell stage using a supervised
single-cell developmental classifier. Features derived from these
classified cell populations—including protein expression levels and
signaling responses under basal and perturbed conditions—were used to
construct a machine learning model (Developmentally Dependent Predictor of
Relapse, or DDPR) that stratifies patients at diagnosis based on future
risk of relapse. This dataset includes raw, bead-normalized ion count
data. It enables exploration of developmental heterogeneity in BCP-ALL and
provides a resource for studying relapse-associated signaling states in
leukemic populations.