dsCellNet

dsCellNet infers cell–cell communication networks from developmental-series RNA-seq data to characterize temporal ligand–receptor interactions and identify active cell types and genes across development and in aging brains.

Key Features:

• Dynamic Time Series Analysis: Handles RNA-seq datasets containing multiple developmental stages to capture temporal changes in cell–cell interactions.
• Protein Localization and Interaction Identification: Identifies potential ligand–receptor interactions based on protein localizations.
• Dynamic Time Warping: Applies dynamic time warping to reinforce inferred interactions within individual time points and across adjacent time points.
• Fuzzy Clustering for Refinement: Uses fuzzy clustering to refine key time points and network connections.
• High Accuracy and Tolerance: Demonstrates high accuracy and tolerance in comparative analyses with other published methods on simulated and real datasets.
• Identification of Active Cell Types and Genes: Pinpoints the most active cell types and genes at various developmental stages.

Scientific Applications:

• Developmental Biology: Maps communication networks across developmental stages to study how cellular interactions drive growth and maturation.
• Aging Research: Analyzes aging brain data to reveal age-related changes in cell–cell communication.
• Disease Mechanism Exploration: Investigates dynamic cell–cell interactions to identify pathways involved in disease progression.

Methodology:

Identifies ligand–receptor interactions using protein localization, applies dynamic time warping across and within time points, refines time points and connections with fuzzy clustering, and evaluates performance via comparative analyses on simulated and real RNA-seq datasets.