scScope

scScope applies deep learning to denoise and identify cell-type composition from millions of noisy single-cell RNA sequencing (RNA-seq) gene-expression profiles.

Key Features:

• Scalability: Processes datasets comprising millions of cells for large-scale single-cell studies.
• Accuracy and Speed: Uses deep learning to rapidly and precisely identify cell types in heterogeneous tissues from noisy gene-expression profiles.
• Handling Dropout Events: Manages dropout events in single-cell RNA-seq by distinguishing technical zeros from biological signal.

Scientific Applications:

• Characterization of cellular states: Enables fine-grained identification of phenotypically distinct cellular states within heterogeneous tissues.
• Tissue and disease studies: Supports analysis of tissue organization and investigations into disease mechanisms and developmental biology.

Methodology:

Deep-learning algorithms are trained on single-cell gene-expression data to learn patterns for denoising, handling dropout events, and rapid classification of cells.