dynwrap: Representing and Inferring Single-Cell Trajectories

Tutorials Reference documentation

dynwrap contains the code for a common model of single-cell trajectories. The package can:

• Wrap the input data of a trajectory inference method, such as expression and prior information
• Run a trajectory inference method in R, in a docker container or a singularity container
• Wrap the output of a trajectory inference method, such as the pseudotime, a clustering or a branch network, and convert it into a common trajectory model
• Further postprocess and adapt the trajectory model, such as labelling the milestones and rooting the trajectory

Documentation and the API reference for dynwrap can be found at the dyvnerse documentation website: https://dynverse.org/ .

dynwrap was used to wrap 50+ trajectory inference method within docker containers in dynmethods.

The advantage of using a common model is that it allows:

• Comparison between a prediction and a gold standard, eg. using dyneval
• Comparing two predictions
• Easily visualise the trajectory, eg. using dynplot
• Extracting relevant features/genes, eg. using dynfeature

Latest changes

Check out news(package = "dynwrap") or NEWS.md for a full list of changes.

Recent changes in dynwrap 1.2.3

• MINOR CHANGE: Fixes to metadata (documentation, citation entry, broken links).

Recent changes in dynwrap 1.2.2

• MAJOR CHANGE convert_milestone_percentages_to_progressions(): Rewrite implementation to attain significant speedup.

• MINOR CHANGE infer_trajectory(): Infer command (Rscript/python) from file extension if possible and avoid using shebang to execute script, because R CMD check for R 4.0 puts Rscript and R dummy executables on the path before R CMD check. This means #!/usr/bin/env Rscript does not work anymore.

• MINOR CHANGE add_feature_importance(): Add a function for adding feature importance scores to a trajectory.

• BUG FIX project_waypoints(): Rename milestone waypoints such that the names are unique.

• BUG FIX infer_trajectory(): Fix error message printing.

• BUG FIX: dyndimred is used conditionally.

• BUG FIX: wrap_expression() and add_expression(): Do not override feature_info when it already exists in dataset.

Dynverse dependencies