dynverse is a collection of R packages aimed at supporting the trajectory inference (TI) community on multiple levels: end-users who want to apply TI on their dataset of interest, and developers who seek to easily quantify the performance of their TI method and compare it to other TI methods.

Some of these packages were developed as part of a benchmarking study ➙

A comparison of single-cell trajectory inference methods

Helena Todorov, Yvan Saeys.

Nat Biotech (Apr. 2019) doi:10.1038/s41587-019-0071-9

bioRxiv (Mar. 2018) doi:10.1101/276907

Want to infer and interpret trajectories?

The dyno package offers end-users a complete TI pipeline. It features:

  • a uniform interface to 50+ TI methods,
  • an interactive guideline tool to help you select the most appropriate method,
  • streamlined interpretation and visualisation of trajectories, and
  • downstream analyses such as the identification of differentially expressed features

Check out the quick start or the user guide.

Need help?

Post an issue on Github ➙

Want to create and evaluate a method?

For developers of existing or new TI methods, dyno offers the same features as to end-users. In addition, developers might also want to check out the following packages:

  • dynmethods, which is a repository of wrappers for TI methods. If your method has already been included in dynmethods, an issue will have been created there.
  • dynwrap, the wrapping functions for transforming common trajectory data formats into the common trajectory model supported by dynverse.
  • dynbenchmark, all source code in order to replicate the benchmarking study by Saelens and Cannoodt (10.1101/276907).
  • Check out this overview of all dynverse packages for more information of the functionality of each package.