# Correlation matrix for biostatistics [closed]

I want to convert the expression of the genes from an array to a gene correlation matrix, to know the correlation of each gene with the other genes. I have 6 samples, 3 controls and 3 test, is it correct to compute the correlation matrix using the 6 samples? Should I consider the 6 samples or just one control and one test?

Thank you!

## closed as unclear what you're asking by Michael Chernick, mdewey, kjetil b halvorsen, gung♦, JohnMay 5 '17 at 17:04

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

• What is correlation of the genes..? What exactly do you measure? – Tim May 1 '17 at 9:07
• I have 6 samples (3 controls and 3 test) and I am measuring how much a specific gene has been expressed. I want to do a correlation matrix to see if there are genes that are always expressed together or the other way around. – Mee May 2 '17 at 12:21

This is a tricky question, because how to define relationship between two genes quite be complicated. To my understanding, bioinformoicans tend to use something like hierarchical clustering, heat-map for comparing genes.

The simplest approach is just take the whole gene expression matrix and apply it to R's cor function. According to:

https://support.bioconductor.org/p/30566/

and

https://www.rdocumentation.org/packages/qpgraph/versions/2.6.1/topics/qpPCC

you may want to use the qpPCC function. Please note:

• All it does is to feed all your samples into the algorithm. The function ignores your sample labels.
• You get a measure of linear correlation between two pairs of genes. But the function doesn't actually know what you mean by "expressing". It doesn't know what pattern you're talking about, all it does is just compare linear correlation between the two genes.