Rüçhan Ekren, PhD
@ruchanekren.bsky.social
Bioinformatics Data Scientist, Immception Lab @infinitytls.bsky.social
https://github.com/rekren
Alumnus of @crctoncopole, @acibademuniv & @iyteedutr
https://github.com/rekren
Alumnus of @crctoncopole, @acibademuniv & @iyteedutr
arxiv.org
October 5, 2025 at 9:45 PM
If its absolutely needed, you can always project the calculated pseudotime on top of any reduced dimension of the data, including UMAP.
November 24, 2024 at 3:18 PM
If its absolutely needed, you can always project the calculated pseudotime on top of any reduced dimension of the data, including UMAP.
Algorithms like Monocle or Slingshot, which infer pseudotime based on distances, may yield erroneous trajectories when the calculation run based on the UMAP coordinates !
Working with higher dimensions (i.e. batch-corrected PCAs) of the data make more sense in terms of data topology.
Working with higher dimensions (i.e. batch-corrected PCAs) of the data make more sense in terms of data topology.
November 24, 2024 at 3:18 PM
Algorithms like Monocle or Slingshot, which infer pseudotime based on distances, may yield erroneous trajectories when the calculation run based on the UMAP coordinates !
Working with higher dimensions (i.e. batch-corrected PCAs) of the data make more sense in terms of data topology.
Working with higher dimensions (i.e. batch-corrected PCAs) of the data make more sense in terms of data topology.