Hi all,

I'm a new user in R , I loaded CEL Files from GEO datasets and I want to analysis my data to detect if there is a Cancer in an experiment sample or not as a purpose of my project , my Question is :

In which a stage of a statistical analysis of DNA microarray could I detect that sample has a tumor  ? using Affymetrix platform .. 

Any one could help please? 

Any advice help me to start my work ?

Cheers,

Tahani.

Your question is quite vague, so I'll give you an equally vague answer; read the limma user's guide.

If you're looking for someone to teach you how to do a DE analysis from start to finish, then you're probably at the wrong place. The Bioconductor support site is just that - it provides support, not teaching. You'd be better off looking at some workflows, like the one linked in an answer to another of your posts (A: help me to analysis my data ). If you've done so and you have specific questions about parts of your analysis, then we'd be in a better position to help you. Otherwise, we'd largely be regurgitating information in the documentation anyway.

Hi,

Sorry for the delay. There is no statistic stage in any genomics data, to predict  if cancer is being or not.

If you would like to compare your data with existing data, you need to clean up the raw data generated by affimetrix. I think there are some packages that do these. I didn't use it.

It is more easier for you to start using more curated data as available at cgdsr  CRAN package.

Biconductor canceR package is a GUI interface, can helps you the be more familiar with  Multi-Assay cancer genomics data.

And then you can start to compare your data with existing data.

Karim