DeepCCI

DeepCCI predicts cell-cell interactions from single-cell RNA sequencing (scRNA-seq) data to map intercellular communication networks.

Key Features:

• Input data: Single-cell RNA sequencing (scRNA-seq) expression matrices from diverse technologies and platforms.
• Modeling approach: Deep learning-based framework that captures complex and latent patterns in scRNA-seq data.
• Sparsity and heterogeneity handling: Designed to address data sparsity and cellular heterogeneity inherent to scRNA-seq.
• CCI prediction: Identifies significant cell-cell interactions between cell types or clusters with high precision.
• Network reconstruction: Constructs comprehensive cell-cell interaction (CCI) networks from predicted interactions.
• Validation: Evaluated across multiple publicly available scRNA-seq datasets obtained from various technologies and platforms.

Scientific Applications:

• Intercellular communication mapping: Mapping communication networks among cell types using scRNA-seq-derived CCIs.
• Cellular differentiation studies: Investigating signaling interactions involved in cellular differentiation.
• Tissue homeostasis analysis: Studying intercellular signaling that maintains tissue homeostasis.
• Immune response research: Analyzing cell-cell communications relevant to immune responses.
• Cross-platform scRNA-seq analysis: Comparative analysis of CCIs across datasets generated by different sequencing technologies and platforms.

Methodology:

Applies deep learning models to scRNA-seq expression data to capture latent patterns and predict cell-cell interactions while addressing data sparsity and cellular heterogeneity; evaluated on multiple publicly available scRNA-seq datasets from different platforms.