Seurat object issue with Ensemble to gene ID conversion for variable genes/plotting after converting from Monocle3 CDS file

[sea-imm-exp]

I am working with another colleague and was provided a monocle3 cds object with integrated samples that have cluster annotations. They do not have the original files, so I need to work with this object for Seurat analyses. The issue I am having is that when I create the seurat object all the genes are ensembl IDs. I created the Seurat object using this code

count.mat <- `assay(obj.cds)`
meta.df <- as.data.frame(colData(obj.cds))

seurat.obj <- CreateSeuratObject(counts = count.mat,
                                project = "new.project",
                                assay = "RNA",
                                meta.data = meta.df)

I have tried a few ways to correct the Ensemble IDs to read in as genes instead (really need this for some plots I'm generating).

1. I created a count matrix from the cds obj and then read in gene symbols. Creating a new Seurat object with the gene annotated count matrix.

2. I have tried creating the seurat object and then either directly replacing the ensembl ID with a gene name or by adding a new metadata column with gene symbols but have been unable to replace var.features or use the genes to label generated graphs.

`seurat.obj@assays[["RNA"]]@meta.data[["var.feats"]]<- ifelse(is.na(gene_symbols$SYMBOL), gene_symbols$ENSEMBL, gene_symbols$SYMBOL)`

I get the following error for both: Error in validObject(.Object) : invalid class “LogMap” object: Duplicate rownames not allowed. Some of the ensembl ID var.features are NA so l think that might be causing some of the issues. I feel like this should be an easy fix because all seurat processing works fine if I keep the ensembl IDs but at this point I'm stuck and would welcome any recommendations. What can I try to correct for duplicate row names or directly substitute out Ensemble IDs for gene names? Or is there a way to keep the gene_short_names from the cds object?