Hay! I am working on an RNA-seq project where they have used a single-cell sequencing technique (SmartSeq3) on conditioned media.
They have sequenced conditioned media from spheroid cultures (~15 samples each from 2 groups, (normal and mutant), across 3 different time points, so around 90 samples), along with ~15 no cell culture and ~5 blank water samples. We wish to compare the RNA profile between normal and mutant.
I processed the data and now have the RNA-seq counts. Since I have the data from no cell culture and water samples, I want to identify and remove their influence from the RNA-seq counts.
I checked RUV-seq, however from what I read and understood, it works best when there are spike-ins, or it will determine a set of "control genes" from samples (which we can select) and use that for batch correction.
Are there any other tools or approaches, like for example, subtracting the median count of genes detected in no cell culture across rest of the samples? How do I make sure that the counts in other 90 samples are not from the media but from the spheroids?
