Gene Names/Identities are showing up as numbers when plotted with highest_expr_genes()

When processing the data in Scanpy I am unable to figure out why my plot of the Highest Expressed Genes shows up with numbers rather than gene names as the identifiers on the Y-axis.
Example:

I am worried that it may not be reading our file properly, but even when I converted from a txt file to a .loom file I still got the same problem. I am unsure if this is an issue with our original input or if something it is being read improperly. I will include some of my code below, and if there is somewhere you would like me to send copies of raw data to run it on let me know where to send it.

I am using the tutorial for Clustering on the Scanpy website as the basis for my code.

sc.settings.set_figure_params(dpi = 80)

day00a = sc.read_text("/alex_ryan/D0.1500.dge", first_column_names = True, delimiter = "\t")
day01 = sc.read_text("/alex_ryan/D1.txt.500.dge", first_column_names = True, delimiter = "\t")
day02 = sc.read_text("/alex_ryan/D2.txt.500.dge", first_column_names = True, delimiter = "\t")
day04 = sc.read_text("/alex_ryan/D4.txt.500.dge", first_column_names = True, delimiter = "\t")
day09 = sc.read_text("/alex_ryan/D9.txt.500.dge", first_column_names = True, delimiter = "\t")
day11 = sc.read_text("/alex_ryan/D11.txt.500.dge", first_column_names = True, delimiter = "\t")

day00a.obs['tech'] = 'Day 0'
day01.obs['tech'] = 'Day 1'
day02.obs['tech'] = 'Day 2'
day04.obs['tech'] = 'Day 4'
day09.obs['tech'] = 'Day 9'
day11.obs['tech'] = 'Day 11'

adata_list = [day01, day02, day04, day09, day11]
adata2 = day00a.concatenate(adata_list, join = 'outer')
adata = adata2

sc.pl.highest_expr_genes(adata, n_top = 20, save = True)

I am adding the extra .obs ‘tech’ tag so that I can identify the cells by day after they have been combined into one anndata object. I don’t think this is causing the issue, but if that is part of it, then please let me know if there is a work around.

[mosquitoCat]

@ivirshup @LuckyMD I fixed the problem - the issue was in the original h5ad file converted from a Seurat object using SeuratDisk::Convert(). It seems the var data wasn’t ported over properly for the assay I was using. I rebuilt the h5ad file using reticulate instead and that solved the problem.

[HKL-97]

@ivirshup @LuckyMD I fixed the problem - the issue was in the original h5ad file converted from a Seurat object using SeuratDisk::Convert(). It seems the var data wasn’t ported over properly for the assay I was using. I rebuilt the h5ad file using reticulate instead and that solved the problem.

I have encountered the same issue as @mosquitoCat . Although adata.var_names still returns correct gene symbols, all my name IDs become numbers: for example, sc.pl.umap(adata,color=‘GeneName’) will return errors. but sc.pl.umap(adata,color=‘123’) can be recognized. SeuratDisk::Convert() seems to cause some trouble here. Is there a way to fix it? @ivirshup