The code has been written in R and I have used the R notebook which makes sharing code and data very easy. I have also made the code as general as possible so that there doesnt need to be much changes, if we would like to use a different dataset or a different set of genes.
The first problem is to arrange and manage the data according to our needs from a public database including changing the variable names.
The second problem is given of a set of expression levels of tumor samples from which we need to classify the different tumor samples into one of the 3 cancer subtypes: squamous, bronchioid, magnoid.
The third problem is to find if the T Cell signature of the 3 cancer subtypes are significantly different from each other or not.
These approaches facilitates us in finding biomarkers of cancer which can be used for genotyping in later studies using simpler technologies such as microarrays.