monocle3-truly-complete

For single-cell trajectory analysis, preprocess scRNA-seq data, reduce dimensionality, cluster cells, learn the developmental graph, and order cells to reconstruct dynamic cellular processes.

Co-embedding and reference projection map multiple scRNA-seq datasets into a common space, enabling label transfer, batch correction, and comparative analysis of new query samples against an existing reference.

Yes, you need an R environment with Monocle 3 (version 1.4.25 or higher) and Bioconductor dependencies, which can be installed via BiocManager or GitHub to execute the trajectory workflows.

Yes, you can project query data onto a reference trajectory to diagnose cellular state transitions, visualize branchpoints, and compare reference trajectories with new samples across development or disease.

The best way to visualize lineage relationships across batches is using co-embedding integration workflows to map datasets into a common space, then applying trajectory learning and pseudotime visualization.

Preprocessing and dimensionality reduction are required before learning the graph because they normalize the scRNA-seq data and extract core structural features, enabling accurate clustering and trajectory construction.