Abstract B-cell acute lymphoblastic leukemia (B-ALL) is the most common childhood cancer. Subtypes within B-ALL are distinguished by characteristic structural variants and mutations, which in some instances strongly correlate with responses to treatment. The World Health Organisation (WHO) recognises seven distinct classifications, or subtypes , as of 2016. However, recent studies have demonstrated that B-ALL can be segmented into 23 subtypes based on a combination of genomic features and gene expression profiles. A method to identify a patient’s subtype would have clear clinical utility. Despite this, no publically available classification methods using RNA-Seq exist for this purpose. Here we present ALLSorts: a publicly available method that uses RNA-Seq data to classify B-ALL samples to 18 known subtypes and five meta-subtypes. ALLSorts is the result of a hierarchical supervised machine learning algorithm applied to a training set of 1223 B-ALL samples aggregated from multiple cohorts. Validation revealed that ALLSorts can accurately attribute samples to subtypes and can attribute multiple subtypes to a sample. Furthermore, when applied to both paediatric and adult cohorts, ALLSorts was able to classify previously undefined samples into subtypes. ALLSorts is available and documented on GitHub ( https://github.com/Oshlack/AllSorts/ ). Key Points ALLSorts is a gene expression classifier for B-cell acute lymphoblastic leukemia, which predicts 18 distinct genomic subtypes - including those designated by the World Health Organisation (WHO) and provisional entities. Trained and validated on over 2300 B-ALL samples, representing each subtype and a variety of clinical features. Correctly identified subtypes in 91% of cases in a held-out dataset and between 82-93% across a newly combined cohort of paediatric and adult samples. ALLSorts assigned subtypes to samples with previously unknown driver events. ALLsorts is an accurate, comprehensive and freely available classification tool that distinguishes subtypes of B-cell acute lymphoblastic leukemia from RNA-sequencing.

ALLSorts: a RNA-Seq classifier for B-Cell Acute Lymphoblastic Leukemia