Requesting assistance with as.Seurat, following Monocle3 Vignette

[BensunCFong]

Hi there,

Long time lurker and Seurat user, first time seeking help here trying to employ Monocle3 with my data.

Background: Generated an integrated Seurat object from two 10X runs (intention differential analysis), and used standard sctransform workflow: anchors, integration, PCA, RunUMAP, FindClusters, FindNeighbors. Then, took a subset of the clusters for object seurat_integrated_2, and repeated PCA, RunUMAP, FindClusters, FindNeighbors.

Following the Monocle3 vignette I can use as.cell_data_set on the object to generate and plot a cell_data_set object with cds <- as.cell_data_set(seurat_integrated_2).

Issue: When I attempt to then convert the object back using integrated.sub <- subset(as.Seurat(cds), monocle3_partitions == 1), I recieve the error message:

Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'x' in selecting a method for function 'subset': One or more of the assays you are trying to convert is not in the SingleCellExperiment object

Specifically running as.Seurat(cds) results in the same error. I can't seem to find any evidence elsewhere online of another user seeing the same issue. This might be an incredibly simple solution, any help would be greatly appreciated. I have the latest versions of Monocle3 and SeuratWrappers installed, and have tried with Seurat 4.0.3 as well as the development release.

[Sa753]

I have the same issue

[torkencz]

Hi, I can take a look at this would you mind giving me an example rds where I can reproduce this and fix it for a future update?