I found that the midway installation of R was having trouble finding important headers when in the conda environment. So we are forced to use an install of R within conda. In this case, it is best practice to install as many packages as we can via conda. I try to only install from github repos within R (although, this also triggers install of any dependencies that are not already installed via conda).
mamba install -c conda-forge r-base r-essentials r-devtools r-reticulate
mamba install -c r r-tidyverse
mamba install -c bioconda r-bioconductor bioconductor-annotationsdbi bioconductor-org.hs.eg.db
mamba install -c bioconda bioconductor-org.hs.eg.db
devtools::install_github("kartayeb/gseasusie@cleanup")