I have scRNA-seq data that I measured based on samples collected at four different time points from both mice with mutations and mice without mutations. Using this data set, I would like to find genes that differ between the presence and absence of mutations at each time point. On the other hand, I would also like to compare the time points only in the group with mutations. To do this, I merged all the data sets using the merge function of Seurat package, then used FindVariableFeatures (nfeatures = 3000) to narrow down the genes that were changing, and used IntegrateLayers function to correct for batch effects. After that, I used the FindMarkers function to search for genes that differ between the presence or absence of mutation at each time point. As a result, I ended up comparing data that included low-expression genes that were almost undetectable at certain time points. Is this approach appropriate? Or should I re-run FindVariableFeatures function, this time limiting it to data from each time point? I would be grateful for your advice.
