Comparing the gene expression data

Question

[Natasha]

I'm trying to compare the gene expression data reported in the studies documented in GEO for a specific gene expressed in a tissue of my interest in Homo sapiens. I compared the values reported in 4 studies, in two studies the expression value of gene A lies between 15-30, in the other two studies the range lies between 1-4.

I would like to ask for advice on whether it is meaningful to compare the data from different studies. What are the factors that one should keep in mind while making such comparisons?

Any help would be highly appreciated.

[Benn]

Hello deepamahm.iisc!

It appears that your post has been cross-posted to another site: https://bioinformatics.stackexchange.com/questions/4891/comparing-the-gene-expression-data

This is typically not recommended as it runs the risk of annoying people in both communities.

[Natasha]

Hello b.nota! Sorry about that

[WouterDeCoster]

You did not mention whether you are looking at RNA-seq or microarray data. "expression value" is also not very informative, it doesn't tell us how that value is determined and if it was normalized. Can you be more specific?

[Natasha]

Sorry for not being clear in the original post. I'm looking at the microarray data. I'm referring to the values reported here

I am not exactly sure about normalization.Could you please explain how to check whether the data is normalized? I'm also attaching a snapshot of the expression value that I obtain from GEO. Please find the link to the snapshot here

[Natasha]

Sorry for not being clear in the original post. I'm looking at the microarray data. I'm referring to the values reported here

I am not exactly sure about normalization.Could you please explain how to check whether the data is normalized? I'm also attaching a snapshot of the expression value that I obtain from GEO.Please find the link to the snapshot

Answer 1

[Santosh Anand]

Usually, the numbers will not mean anything across the studies, unless you normalize them together. But then, if the data is coming from heterogeneous platforms, the normalization is not straightforward - and you have to also account for batch effect, different studies, different designs etc etc...

The other thing you can do is "internal normalization". For example, if you know a gene that should not vary its expression across studies ("sort of a housekeeping gene"), you can get a fold change compared to that gene in each sample, and compare the fold changes. That would be more robust.

[Natasha]

Thank you very much for the explanation. I want to compare the gene expression data from different studies for all the genes involved in 5 metabolic pathways. Could you please explain how to get "a fold change compared to that gene in each sample, and compare the fold changes"? I'm really sorry for the naive questions.I'm new to this field.