FateID

FateID quantifies and visualizes fate bias in multi-lineage single-cell transcriptome datasets to resolve lineage commitment and derive differentiation trajectories.

Key Features:

• Probabilistic quantification of cell fate bias: FateID employs an iterative supervised learning algorithm to probabilistically quantify fate bias in progenitor populations and detect subtle transcriptome differences indicative of lineage bias.
• High-resolution differentiation trajectories: FateID delineates domains of fate bias to derive high-resolution differentiation trajectories that track progenitor evolution and lineage commitment.
• Integration with RaceID3 clustering: FateID integrates with the RaceID3 clustering algorithm to improve cell-type identification and enable fine-grained lineage analysis.

Scientific Applications:

• Stem cell research: FateID maps differentiation pathways and fate choices of progenitor cells from single-cell transcriptome data.
• Hematopoietic progenitor analysis: Applied to mouse hematopoietic progenitors enriched for lymphoid lineages, FateID identified lineage relationships including a common progenitor population of B cells and plasmacytoid dendritic cells.
• Experimental validation: FateID predictions can be validated using in vitro differentiation assays.

Methodology:

FateID uses an iterative supervised learning algorithm and integrates with the RaceID3 clustering algorithm; it was developed alongside a robotic miniaturized CEL-Seq2 implementation for deep single-cell RNA sequencing of approximately 2,000 mouse hematopoietic progenitors.