Deep learning-based classifier for malignant plasma cell identification in myeloma
4 affiliations
- 1Department of Pediatrics, Emory University, Atlanta, GA USA
- 2Department of Biomedical Informatics, Emory University, Atlanta, GA USA
- 3Coulter Department of Biomedical Engineering, Emory University, Atlanta, GA USA
- 4Aflac Cancer and Blood Disorders Center, Children’s Healthcare of Atlanta, Atlanta, GA USA
Abstract
Multiple myeloma (MM) displays significant genetic heterogeneity, making it challenging to distinguish malignant from non-malignant plasma cells in single-cell datasets. Existing marker-based and CNV detection methods require manual intervention or high computational resources, limiting their scope. We developed a supervised deep learning autoencoder to classify malignant cells across MM and its precursor stages and validated its performance and biological relevance. The model outperformed alternative approaches, achieving mean AUCs of 0.86 on internal and 0.80 on external datasets, with strong performance on unseen samples across single-cell platforms (mean AUC 0.92). Despite training solely on MM samples, our model distinguishes malignant cells in precursor stages, with predicted malignant cell proportions increasing from monoclonal gammopathy of undetermined significance (MGUS, 29.00%) to smoldering multiple myeloma (SMM, 77.83%) and MM (96.08%). In paired samples, the model also accurately captured a reduction in malignant plasma cell proportions following treatment, reflecting patient variability in treatment response. Differential expression analysis uncovered a 12-gene malignant plasma signature linked to poor survival (HR = 2.7, Log-rank P = 0.0023) and validated in the MMRF CoMMpass cohort. Overall, our resulting model enables accurate malignant cell identification and provides biologically and clinically relevant insights for MM research.
Related
- Differentiating benign from malignant adnexal masses by biomarker-agnostic plasma proteomics using adaptive machine learning
- A machine learning classifier reveals hidden rare biallelic heterozygous variants in the coding regions of Meniere disease genome
- Stage-Stratified Explainable Machine Learning Identifies a Reproducible Transcriptomic Classifier with Stage-Dependent Feature Importance in Colorectal Cancer: A Multi-Cohort Study with Saudi Population Contextualisation
- The Converging Roles of DNA Damage Resolution and Redox Dysregulation in Plasma Cell Transformation During the Evolution of Multiple Myeloma.
- FRANK: a pan-cancer RNA-seq classifier for childhood tumors with data-efficient learning
BibTeX
@article{satpathy2026deep,
title = {{Deep learning-based classifier for malignant plasma cell identification in myeloma}},
author = {Sarthak Satpathy and Marina E. Michaud and William C. Pilcher and Swati S. Bhasin and Manoj K. Bhasin},
journal = {NPJ Precision Oncology},
year = {2026},
month = jul,
doi = {10.1038/s41698-026-01589-6},
url = {https://doi.org/10.1038/s41698-026-01589-6},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Deep learning-based classifier for malignant plasma cell identification in myeloma
AU - Sarthak Satpathy
AU - Marina E. Michaud
AU - William C. Pilcher
AU - Swati S. Bhasin
AU - Manoj K. Bhasin
PY - 2026
DA - 2026/07/09
JO - NPJ Precision Oncology
DO - 10.1038/s41698-026-01589-6
UR - https://doi.org/10.1038/s41698-026-01589-6
AB - Multiple myeloma (MM) displays significant genetic heterogeneity, making it challenging to distinguish malignant from non-malignant plasma cells in single-cell datasets. Existing marker-based and CNV detection methods require manual intervention or high computational resources, limiting their scope. We developed a supervised deep learning autoencoder to classify malignant cells across MM and its precursor stages and validated its performance and biological relevance. The model outperformed alternative approaches, achieving mean AUCs of 0.86 on internal and 0.80 on external datasets, with strong performance on unseen samples across single-cell platforms (mean AUC 0.92). Despite training solely on MM samples, our model distinguishes malignant cells in precursor stages, with predicted malignant cell proportions increasing from monoclonal gammopathy of undetermined significance (MGUS, 29.00%) to smoldering multiple myeloma (SMM, 77.83%) and MM (96.08%). In paired samples, the model also accurately captured a reduction in malignant plasma cell proportions following treatment, reflecting patient variability in treatment response. Differential expression analysis uncovered a 12-gene malignant plasma signature linked to poor survival (HR = 2.7, Log-rank P = 0.0023) and validated in the MMRF CoMMpass cohort. Overall, our resulting model enables accurate malignant cell identification and provides biologically and clinically relevant insights for MM research.
SN - 2397-768X
ER -