Gene ID conversions

[annaborchers]

Hi there - I'm using some scRNAseq data that only has the Ensemble IDs, and when I run with Azimuth it goes through, however I noted this detail as it was running:

"detected inputs from HUMAN with id type Gene.stable.ID
reference rownames detected HUMAN with id type Gene.name
Found 17658 out of 36601 total inputs in conversion table
Normalizing query using reference SCT model
Warning: No layers found matching search pattern provided
Warning: 1386 features of the features specified were not present in both the reference query assays.
Continuing with remaining 3614 features."

My interpretation is that it is now only using a very limited number of genes to identify my clusters, however I would of course prefer to maximize the accuracy by using as many as possible. Anybody know how to best approach this?

Thanks!