Leveraging single-cell sequencing to classify and characterize tumor subgroups in bulk RNA-sequencing data.

Area of Science:

• Computational biology
• Genomics
• Cancer research

Background:

• Accurate cancer subtyping is crucial for precision medicine.
• High-throughput sequencing generates large-scale transcriptomic data for cancer analysis.
• Computational methods can improve cancer subtyping and personalize treatments.

Purpose of the Study:

• To evaluate feature selection schemes for meningioma classification.
• To develop an algorithm integrating bulk and single-cell data for improved cancer subtyping.
• To enhance personalized treatment strategies through better cancer diagnoses.

Main Methods:

• Developed CLIPPR algorithm to combine single-cell models, RNA-inferred copy number variation (CNV) signals, and bulk models.
• Evaluated feature selection schemes for meningioma classification using bulk (77 samples) and single-cell (10K cells) data.
• Validated the algorithm on 789 bulk meningioma samples from multiple institutions and TCGA glioma data (711 samples).

Main Results:

• Bulk transcriptomic data showed good accuracy but confused benign and malignant classes (8% of samples).
• Single-cell data resolved confused subgroups but had limited overall accuracy.
• CLIPPR demonstrated superior overall accuracy and resolved benign-malignant confusion.

Conclusions:

• CLIPPR synergizes single-cell resolution with bulk sequencing depth for improved cancer subtyping.
• The algorithm enhances cancer subgroup diagnoses and provides biological insights.
• CLIPPR offers a powerful tool for advancing precision oncology.