search

constantAmateur / SoupX

R package to quantify and remove cell free mRNAs from droplet based scRNA-seq data
249 stars 34 forks source link

How to set the contamination range for function autoEstCont? #65

Closed honghh2018 closed 3 years ago

honghh2018 commented 3 years ago

Hi developer, Thanks for the nice tools to process ambient mRNA. Even though the process and vignitte is clear for me, the contaminationRange of autoEstCont's parameter had me confused. how to set the contaminationRange for my own data if i consider the ambient rate of my 10X scRNA data as 2%-5% or so. What mean the contaminationRange default value between 0.01 and 0.8 ? Any kindly help would be appreciated. Best,

constantAmateur commented 3 years ago

This parameter just sets the range of plausible values for the contamination fraction. A solution will only be considered valid if it falls within this range.

For most datasets, you could set this to c(0,1) (equivalent to no constraint) and it wouldn't make a difference. Occasionally, datasets can have a big spike of estimates at exactly 0, which the defaults prevent being chosen as the preferred solution.

Basically, it's fine to ignore it and use the defaults. Unless your data produces some weird multi-modal distribution of estimates of the contamination when you run autoEstCont, in which case you can use this parameter to filter out parts of the parameter space for the contamination fraction.