CellRank: consistent and data view agnostic fate mapping for single-cell genomics.

Area of Science:

• Computational Biology
• Genomics
• Molecular Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) provides high-resolution insights into cellular differentiation.
• Computational reconstruction of cellular trajectories is crucial due to the destructive nature of scRNA-seq experiments.
• Existing methods like the initial CellRank iteration inferred trajectories from RNA velocity and transcriptomic similarity.

Purpose of the Study:

• To generalize the CellRank trajectory inference framework for multiview single-cell data.
• To develop a scalable framework for comprehensive cellular fate mapping.
• To enable the consistent quantification and analysis of cellular fate and lineage priming.

Main Methods:

• Development of CellRank 2, a generalized computational framework for trajectory inference.
• Integration of complementary data views (e.g., experimental time points, pseudotime, stemness potential) into the analysis.
• Construction of cell-cell transition matrices to induce Markov chains for inferring terminal states and lineage formation.

Main Results:

• CellRank 2 successfully generalizes trajectory inference to multiview single-cell data.
• The framework enables scalable and consistent quantification of cellular fate.
• Demonstrated ability to combine complementary data views for a more holistic lineage analysis.

Conclusions:

• CellRank 2 offers a powerful and versatile tool for cellular fate mapping using scRNA-seq data.
• The framework facilitates a deeper understanding of biological differentiation and disease progression.
• Provides protocols for scalable CellRank analyses across diverse data views, requiring Python proficiency.