How do I get estimated gene counts for a reference that contains GFP using Kallisto?

Question

[bioinfo]

Hello,

I am interested in aligning some samples to a human-GFP reference using kallisto. From what I understood to make the new kallisto reference I just need to concatenate the human cDNA fasta file with the GFP fasta file and then create the index with kallisto. However, I am not sure how to produce the gene counts afterwards. Normally, following allignment I use tximport and the ensembl package (through Biomart) to generate counts. Should I add GFP in the tx2gene dataframe that contains the transcript and gene IDs?

Thank you