Running Monocle with cell-v-cell distance as an input

[k3yavi]

Hi, I have some set of data which needs to be batch corrected before making time-series plots. I am thinking of using Seurat for batch correction upstream of Monocle but due to the follow issue it does not report expression values after batch correction. I am wondering is their any defined procedure for Monocle to be used for batch correction? If not, can Monocle take the cell-v-cell distance as an input to make the time-series plots as suggested by Seurat's author (as in the issue linked above).

Thanks!

[Xiaojieqiu]

in theory, it is possible to use cell-v-cell distance to do trajectory inference. however, this functionality is not support in monocle. This is because as datasets becoming larger, a cell-cell matrix will cost large memory footprint and is thus not a good design decision. I suggest you to manually performing batch correction (if that is really what you want) and create a cds based on the corrected data for trajectory inference. Note also that monocle 2 can perform batch corrections too with residualModelFormulaStr argument in reduceDimension function

[k3yavi]

Thanks for the reply @Xiaojieqiu

[Xiaojieqiu]

you are welcome. please let me know if you have any further questions!