How to predict CNV based on Gene Expression dataset

Question

[Kritika]

Hi ,

I have Microarray dataset of Gene expression. I want to predict CNV based on this Microarray Gene expression dataset. Is there any possibility to do this?. Is there any tools available which can help me.?

[Ram]

I've seen CNV processing based on high throughput sequencing and genotyping arrays, but doesn't Microarray tackle RNA expression? Do you think it will be a reliable dataset for Copy Number Variants? How would one deconvolute natural differential expression from DE resulting from a duplication/deletion?

[Kevin Blighe]

There are many microarrays that specifically target copy number. Also, there is a long-running project, called AROMA, that has developed methods to derive copy number from SNP arrays. In this case, the user is indeed referring to a cDNA microarray that aims to measure gene expression.

Answer 1

[h.mon]

To be able to reliably predict CNVs from gene expression data, a requirement would be a more or less simple and predictable relation between copy number and gene expression variations. However, it seems this is not the case: copy number and gene expression have a complex (and not fully understood) relationship. See the review The impact of human copy number variation on gene expression, for example.