Related Experiment Video
Updated: Jul 17, 2025

14:06
Isolation of Precursor B-cell Subsets from Umbilical Cord Blood
Published on: April 16, 2013
18.1K
The Gene Expression Classifier ALLCatchR Identifies B-cell Precursor ALL Subtypes and Underlying Developmental
Thomas Beder1, Björn-Thore Hansen1, Alina M Hartmann1,2
1Medical Department II, Hematology and Oncology, University Hospital Schleswig-Holstein, Kiel, Germany.
Hemasphere
|August 30, 2023
Summary
ALLCatchR, a machine learning tool, accurately classifies B-cell precursor acute lymphoblastic leukemia (BCP-ALL) subtypes using RNA-Seq gene expression data. This advances diagnostic capabilities for BCP-ALL, offering insights into developmental trajectories.
Area of Science:
- Computational biology and bioinformatics
- Hematology and oncology
- Molecular diagnostics
Background:
- Current B-cell precursor acute lymphoblastic leukemia (BCP-ALL) classifications rely on genomic drivers, but systematic gene expression analysis is less advanced.
- Transcriptome sequencing (RNA-Seq) is established for identifying driver aberrations, yet robust methods for gene expression-based subtyping are needed.
Purpose of the Study:
- To develop ALLCatchR, a machine learning classifier for accurate molecular subtyping of BCP-ALL using RNA-Seq gene expression data.
- To enable routine application of RNA-Seq for BCP-ALL diagnostics and provide novel insights into developmental trajectories.
Main Methods:
- Developed ALLCatchR, a machine learning classifier trained on 1869 BCP-ALL transcriptome profiles across four cohorts.
- Validated ALLCatchR on three independent hold-out cohorts (n=1018), assessing accuracy, sensitivity, and specificity for subtype allocation.
- Established a novel RNA-Seq reference of human B-lymphopoiesis from FACS-sorted healthy bone marrow progenitor stages.
Main Results:
- ALLCatchR achieved 95.7% overall accuracy in subtype allocation across independent cohorts, with high-confidence predictions in 83.7% of samples.
- The classifier outperformed existing tools and identified novel driver candidates, with only 1.2% of samples remaining unclassified.
- Projection of BCP-ALL samples onto the normal lymphopoiesis trajectory revealed shared proximity patterns, offering a new framework for developmental comparisons.
Conclusions:
- ALLCatchR provides a systematic and accurate RNA-Seq-based approach for BCP-ALL molecular subtyping, enhancing diagnostic capabilities.
- The tool facilitates routine RNA-Seq application in BCP-ALL diagnostics, improving subtype allocation and providing novel insights into disease biology.
- The integration of a normal lymphopoiesis reference trajectory offers a framework for understanding BCP-ALL developmental origins and classifications.
Related Concept Videos
Differentiation of Common Myeloid Progenitor Cells
3.2K
Common myeloid progenitors (CMPs) are oligopotent cells that can differentiate into granulocytes and macrophages. Granulocytes and macrophages are essential for protecting the body against bacterial, viral, or fungal infections. They migrate from the bone marrow into the circulating blood to reach specific tissue sites where they differentiate and help in immune surveillance. However, they survive only for a few days and must be continuously made available to the organism to maintain a robust...
3.2K
B Cell Activation and Differentiation
1.8K
The adaptive immune response, a sophisticated defense mechanism, relies on the activation and differentiation of B lymphocytes, or B cells. These processes enable our bodies to mount a tailored response against specific pathogens such as bacteria, free virus particles, toxins, and parasites.
When naive B cells encounter a specific antigen that can bind to the B cell receptor (BCR) on their surface, they undergo sensitization to respond to the antigen's presence. Sensitization begins with...
When naive B cells encounter a specific antigen that can bind to the B cell receptor (BCR) on their surface, they undergo sensitization to respond to the antigen's presence. Sensitization begins with...
1.8K

