Statistical method for evaluating gene expression

#1
Hey guys,

I am trying to see how one gene affect other gene expression in cancer. I am going to be using microarray data to obtain mrna, mutation and copy number data and then I will analyse it against other genes mrna data to see if they are downregulated or upregulated in presence of the first genes mutations, mrna and copy number changes. I have 4 datasets with 30 genes each, so quite a lot of work for a short period of time. I havent decided which stats software to use as im not overly confident in stats and this grade is really important for my masters. I also have 6 different mutations for that gene and I would like to analyse how each of them affects my other genes individually. Any ideas what tests would be best to use? also, do i need to normalize gene expression data if I am using non parametric tests like Mann Whitney? i was thinking of doing log2 transformation for that? and also drawing q-q plot to show that data is definitely normally distributed? other test option from mann whitney were also one way anova, kruskall wallis, etc. any idea which one would be most appropriate? also, anything else i should do to attract a high mark from this? thank you for the help and sorry for the long post