Error running Monocle3 about seurat-wrappers

[mojaveazure]

One of the requirements of Monocle 3 is that the dimensional reduction names are all upper-case, while Seurat defaults to lower-case names (eg. "UMAP" vs "umap"). The converter does not yet take care of this, but we are working on it. In the meantime, you can either pass reduction.name = "UMAP" to RunUMAP (as seen in the vignette) or rename a previously calculated UMAP result in Seurat before converting to Monocle 3

seurat.object[["UMAP"]] <- seurat.object[["umap"]]
# To save on memory, you can delete the original UMAP, but this is not necessary
seurat.object[["umap"]] <- NULL

Please note, the same issue applies for PCA (Seurat stores as "pca" instead of "PCA") and the same fix for UMAP applies here as well.

[bensong721]

Just wanted to say that I too am having this issue.

cds <- as.cell_data_set(ciona.integrated)
Warning message:
“Monocle 3 trajectories require cluster partitions, which Seurat does not calculate. Please run 'cluster_cells' on your cell_data_set object”

cds <- cluster_cells(cds)
Error: No dimensionality reduction for UMAP calculated. Please run reduce_dimension with reduction_method = UMAP before running cluster_cells

cds <- reduce_dimension(cds, reduction_method = "UMAP")
No preprocess_method specified, using preprocess_method = 'PCA'

cds <- preprocess_cds(cds, method = "PCA")
Error in names(sf) <- colnames(SingleCellExperiment::counts(cds)): attempt to set an attribute on NULL

Any help would be appreciated!

[ashuchawla]

Wondering if you guys ran UMAP on Seurat object at all?

integrated <- RunUMAP(integrated, dims = 1:<num_of_PCs>, reduction.name = "UMAP")

[ashwinikumarkulkarni]

@ashuchawla Yes! RunUMAP was already run during Seurat Integration pipeline. I already have the UMAP plot. And I want to map Monocle's pseudotime onto same UMAP coordinates.. which is where the error is.

[mojaveazure]

This issue should be resolved in the latest version of SeuratWrappers. Thanks for reporting!

[ashwinikumarkulkarni]

Thanks a lot @mojaveazure! I think this will work. I am trying now!