Error in merging datasets: Error in irlba

[atermanini]

Hi,

I am trying to merge datasets, my code follow the tutorial and worked fine with other datasets, but with those samples I got this error:

Error in irlba(A = t(x = object), nv = npcs, ...) : 
  starting vector near the null space
Calls: <Anonymous> ... RunIntegration -> RunPCA -> RunPCA.default -> irlba

Anyone can help me?

I am using the last Seurat v3 from develop branch.

Thanks very much, Alberto

Merging dataset 7 into 2
Extracting anchors for merged samples
Finding integration vectors
Finding integration vector weights
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
Integrating data
Merging dataset 2 7 into 5
Extracting anchors for merged samples
Finding integration vectors
Warning in irlba(A = t(x = object), nv = npcs, ...) :
  fast code path error starting vector near the null space; re-trying with fastpath=FALSE.
Error in irlba(A = t(x = object), nv = npcs, ...) : 
  starting vector near the null space
Calls: <Anonymous> ... RunIntegration -> RunPCA -> RunPCA.default -> irlba
Inoltre: Warning messages:
1: In CheckDuplicateCellNames(object.list = object.list) :
  Some cell names are duplicated across objects provided. Renaming to enforce unique cell names.
2: In `[<-.data.frame`(`*tmp*`, , 2, value = list(1126L, 1134L, 1135L,  :
  provided 8463 variables to replace 1 variables
3: In `[<-.data.frame`(`*tmp*`, , 2, value = list(1777L, 1777L, 1778L,  :
  provided 19214 variables to replace 1 variables
Esecuzione interrotta

[dviraran]

Hi,

I have no idea. This is something related to Seurat and not SingleR. Try creating Seurat objects of these samples and merge them, maybe you can detect the issue.

Best, Dvir