RunVelocity() uses "data" rather than "counts" matrix about seurat-wrappers HOT 2 OPEN

New to velocity analysis. Thanks for this note! I'm wondering, in terms of the correctness, should we use the raw counts or some normalized values for RunVelocity()?

Thanks,
Yichao

from seurat-wrappers.

Hi !
@krejciadam, if you execute line by line the code contained in RunVelocity, you will see that before calling velocyto.R::gene.relative.velocity.estimates, the function creates a list containing the data (spliced-emat/unspliced-nmat) and the argument to pass to velocyto (deltaT, ...)

names(args)
[1] "emat" "nmat"
[3] "deltaT" "steady.state.cells"
[5] "kCells" "kGenes"
[7] "old.fit" "mult"
[9] "min.nmat.smat.correlation" "min.nmat.emat.correlation"
[11] "min.nmat.emat.slope" "zero.offset"
[13] "deltaT2" "fit.quantile"
[15] "diagonal.quantiles" "show.gene"
[17] "do.par" "cell.dist"
[19] "emat.size" "nmat.size"
[21] "cell.emb" "cell.colors"
[23] "expression.gradient" "residual.gradient"
[25] "n.cores" "verbose"

If you look at the distribution of values of emat and nmat, you will see that they are identical to the distribution of values you have in your loom files

summary(args[["emat"]][,1])
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 0.000 3.929 0.000 529.000
summary(args[["nmat"]][,1])
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 0.000 1.268 0.000 265.000
summary(loom$exon[,1])
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 0.000 0.000 2.509 0.000 529.000
summary(loom$intron[,1])
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.0000 0.0000 0.0000 0.8518 0.0000 265.0000

Therefore, it seems pretty clear for me that RunVelocity provides raw counts to velocyto.R

Best regards

from seurat-wrappers.