Can microarray and RNA Seq. gene expression values from the same dataset be compared across genes? Can they be used for numeric calculations? If the expression of one gene is twice as high as another, can I then conclude that this gene is twice as much transcribed?
Hi Thomas,
This question has been answered many a time. If you search biostars or when you enter the question into the search box, similar questions like yours will populate. So instead of making a redundant question and wait for someone to reply, you can easily click on a previous post that will also answer your question.
First, you should know that qPCR is much more sensitive than other expression technologies, and such is the case when validating microarray and RNA-seq expressions. Likewise, RNA-seq has a broader dynamic range compared to microarray. Next, yes, microarray and RNA-seq can be compared even if it's not the same dataset. You have to put both expression types on the same scale. You can do this by converting one expression type or taking the log2 of the expression ratio. Think about the MA plot!! Anyhow, see How To Go About Comparing Rnaseq With Micro Array Expression Data?
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
What type of microarray and normalization methods? For oligo microarrays (e.g. Affymetrix) - generally no. Hybridization intensities can be probe-sequence dependent, so if Gene A has an intensity of 10 and Gene B has intensity 20, it would be very risky to to conclude B is expressed at twice the level of A. Broadly there tends to be a relationship between intensity and abundance across genes, but it is noisy. For RNASeq if you use a summary method that accounts for the transcript lengths of A and B, I'd guess you would be safer in making that conclusion, but others might have more informed advice.