RNA Seq Classification

Question

[mosquitoking]

I have a data set of RNA-Seq stored in an excel file. The file contains gene expression values of two groups, control and disease. I am wondering if it is possible to, given the current data, categorize future genes into either control or disease based on their levels of gene expression and p value.

I am new to machine learning algorithms and have not delved deep into any software's or languages yet, although I'm considering learning R and/or Octave to help me with my data sets.

[seidel]

You might be mixing up a few things....this statement in particular: "...categorize future genes into either control or disease based on their levels of gene expression and p value." What do you mean when you say "future genes"? And what would the p-value reflect? If you have a p-value, then you already have a hypothesis for that gene, i.e. The gene is not affected by the disease state (this is the null hypothesis), and your p-value would allow you to discard that hypothesis in favor of the alternate hypothesis (the gene is affected by the disease state). Remember that each gene generates its own hypothesis in an RNA Seq experiment. Your excel file has only two groups? Were replicates in each group performed? Perhaps you can explain a little more, and in particular what machine learning has to do with any of this.