Hello Biostars!

Let's assume that I have the 2 allelic versions of a given gene (Gene A). To be precise, I have previously characterised a heterozygous SNP on Gene A, and I've generated the 2 possible allele versions:

1. Reference Gene A: reference sequence (the SNP is not applied)

2. SNP Gene A: reference sequence but with SNP change applied.

Let's assume too that I've generated the corresponding transcript sequences for both 1. and 2. Now, I would like to measure the expression of both of them. In order to do this, I am planing to use kallisto in the following way:

• Download reference transcriptome fasta file from ensembl (as recommended by kallisto developer).
• Delete Gene A from reference transcriptome fasta file and add my sequences 1. and 2. (yes I know, 1. is equal to the reference that is already included in the reference transcriptome so, what I am doing basically is adding SNP Gene A sequence to the transcriptome fasta file).
• Run Kallisto using the transcriptome that includes both alleles of the Gene A.

• Is this approach sensible?
• Would the TPMs obtained for my 2 alleles be "comparable" with the TPMs of the rest of the transcriptome (reference)?
• Maybe you can recommend a better approach?

Thank you!