search

LiQian-XC / sctour

A deep learning architecture for robust inference and accurate prediction of cellular dynamics
https://sctour.readthedocs.io
MIT License
51 stars 4 forks source link

Trajectory inference of datasets with batch #2

Closed wu-yc closed 2 years ago

wu-yc commented 2 years ago

Hi Qian,

Thanks again for this beautiful work.

I have lists of UMI matrix, integrated them by using scTransform or harmony (in R, and transfer it to h5ad), and generated the dimension reduction coordinates (i.e. UMAP). Those lists of UMI matrix are generated at different timepoints and treatment conditions. When following the notebook of _scTour_inferencebasic.ipynb, it seems that the inferred trajectory is from batch 1 to batch 2 cells.

Cheers, Yingcheng

LiQian-XC commented 2 years ago

@wu-yc did you use the corrected counts as the input for scTour? I suggest to use the raw UMI count matrix from Cellranger as the input. Do you mind pasting your code here? It will be easier for me to check from there. Thanks!

wu-yc commented 2 years ago

Thanks for your kind support. Now i used the pre-computed UMAP (from Seurat) and then calculate the ptime. It finally works beautifully.