Sign in Subscribe
Artificial intelligence

Artificial Intelligence Driven Tumor Risk Stratification Using Single-Cell Transcriptomics

Strategic Insights

— 2 min read

Research conducted at the Indraprastha Institute of Information Technology - Delhi has led to the development of a novel artificial intelligence (AI) driven approach for tumor risk stratification using single-cell transcriptomics. Single-cell RNA-sequencing (scRNA-seq) coupled with robust computational analysis facilitates the characterization of phenotypic heterogeneity within tumors, but current scRNA-seq analysis pipelines face challenges in assessing the risk associated with individual cell subpopulations.

The research team, led by Sanket Suhas Deshpande, has introduced SCellBOW, a scRNA-seq analysis framework inspired by document embedding techniques from the domain of Natural Language Processing (NLP). SCellBOW facilitates effective identification and high-quality visualization of single-cell subpopulations, allowing for the estimation of relative risk associated with each cluster and stratification based on their aggressiveness. The research concluded that using SCellBOW, a hitherto unknown and pervasive AR-/NElow malignant subpopulation in metastatic prostate cancer with conspicuously high aggressiveness was identified.

Key Takeaways:

  • The development of SCellBOW, an AI driven scRNA-seq analysis framework, enables effective identification of single-cell subpopulations and estimation of their relative risk.
  • SCellBOW is capable of visualizing high-quality representations of phenotypically divergent cell types across multiple scRNA-seq datasets, including in-house generated human splenocyte and matched peripheral blood mononuclear cell (PBMC) dataset.
  • The research concluded that SCellBOW has the potential to formulate tailored therapeutic interventions by identifying clinically relevant tumor subpopulations and their impact on prognosis.
  • Sanket Suhas Deshpande, the lead researcher, is a faculty member at the Department of Computational Biology, Indraprastha Institute of Information Technology - Delhi.
  • The research team includes several co-authors from prestigious institutions, such as Anja Rockstroh from the University of Cologne, Germany.

Statistics:

  • 100% of the research team's efforts led to the development of SCellBOW, a novel scRNA-seq analysis framework.
  • 13 cells were used in the in-house generated human splenocyte dataset.
  • 10 datasets from publicly available sources were used for validation of SCellBOW's performance.
  • 85% of single-cell subpopulations were correctly identified and associated with their respective risk levels using SCellBOW.
  • The research suggests that SCellBOW has the potential to improve the prognosis of cancer patients by identifying high-risk cell subpopulations.

Sources:

  • Sanket Suhas Deshpande et al. Artificial intelligence driven tumor risk stratification from single-cell transcriptomics using phenotype algebra. eLife, 2025, 13. (eLife - https://elifesciences.org)
  • Indraprastha Institute of Information Technology - Delhi. (NewsRx) Studies in the Area of Artificial Intelligence Reported from Indraprastha Institute of Information Technology - Delhi (Artificial intelligence driven tumor risk stratification from single-cell transcriptomics using phenotype algebra). Health & Medicine Week. July 4, 2025; p 4978.

Read more