I agree that figuring out how to normalize gene copy is critical. Even the standard denominators can be troublesome as you have mentioned, and I had a similar challenge with a variable water % sample that was then normalized to extracted DNA even though that was also expected to be variable in samples but at least less variable than sample weight.
I would recommend evaluating during development in the context of the overall variability of the readout. If you have two or three sources of analytical error with or without tissue composition error, what is the total error in the reported value? Is it better or worse than the total error of the original denominator. If each readout is 20% error, and the tissue composition is 50% error, then it may be worth correcting but less likely if the readout is 30% error and the tissue composition is 40% error. So consider propagating the error in your calculations before going too far down the road of adding assay complexity.
------------------------------
Joleen White Ph.D.
AAPS 2024 Global Health Community Chair
Bioanalytical 101 Course Development
Senior Advisor
BioData Solutions LLC
[email protected]Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer.
------------------------------
Original Message:
Sent: 06-03-2025 10:08
From: Michael Salgado
Subject: Gene expression mRNA copy number determination by qPCR
I'm wondering if for gene expression assays via qPCR if this community has any recommendations and thoughts on possibly using a relative mRNA copy number calculation. This is in relation to trying to assess gene expression from a small target organ which when excised contains varying portions of other surrounding tissue. Currently, an assay for the gene of interest is duplexed with a housekeeping gene assay which is only used as a control. The mRNA copy number is only back calculated from the gene of interest assay. I'm trying to think in terms of a delta Ct or delta delta Ct like for gene expression ratios.
In one option I see a normalization factor calculated based on a set of control sample Cts of cleanly prepared organ(little surrounding tissue) with a gene specific to the organ of interest and the Ct of that same assay in the sample. This normalization factor could then be used to transform the Ct value of the assay for the gene of interest. In this case it seems that there is no accounting in the normalization for RNA quality/individual sample variability.
In another option I see including a ubiquitous housekeeping gene assay, a gene specific assay to the organ of interest, and an assay specific to the gene of interest being expressed in the organ in a triplexed assay. The normalization factor would be calculated using the Ct of the ubiquitous housekeeping gene assay and the Ct of the gene specific assay of the organ of interest. This normalization factor could then be used to transform the Ct value of the assay for the gene of interest in the reaction. This normalization seems to take into account RNA quality and individual sample variability.
I recognize overall that there are flaws with both options. This then becomes a relative mRNA copy number as the original back calculated value will possibly be increased or decreased after normalization. Ultimately, the assay specific to the gene of interest should be detecting the mRNA if the tissue is excised, handled(flash freeze and -80 storage), extracted, and tested(Reverse transcription and qPCR) properly. The back calculated value of the mRNA copy number should be correlative to the starting tissue amount if expression is occurring regardless of background RNA from surrounding tissues. Curious if anyone has worked with challenging gene expression assays and tissues.
------------------------------
Michael Salgado
Scientist
Regeneron
Tarrytown NY
[email protected]
Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer.
------------------------------