Starlng: STability Analysis of coReguLated "Nests" of Genes.

Starlng identifies stable, robust gene modules (“nests”) by filtering outlier clusters using pseudotime variance, improving covariation-based gene graph clustering for developmental trajectories.

• Why it matters: Gene clustering for single-cell state discovery often fails to capture spatial organization and relationship dynamics along trajectories while remaining sensitive to noise and stochasticity.
• What they did: Building on ClustAssess, Starlng provides an optimized interactive R pipeline that finds stable clustering parameter configurations and robust gene modules, then removes outlier clusters based on pseudotime variance.
• The result: Applied to single-cell data, Starlng yields reliable, reproducible gene communities and reveals unbiased hierarchical relationships between higher-level cell clusters and more granular gene subgroups.