3. Consider the SRBCT microarray data. (X and Y are in different files. Note the data does not provide column names, so when reading into R using read.csv, specify read.csv("X_SRBCT.csv", header=F".)) This is a gene expression data set from a childhood cancer study with n=83 patients and p=2308 genes. The response is the expression profile of p53, a major oncogene that acts as a tumor suppressor. The goal is to select a relatively smaller number of other genes whose expression profiles are associated with p53. a) Among the regression methods we learned, including least square, subset selection, ridge regression, principal component regression, and the lasso, which one(s) is (are) appropriate? Explain. b) Visualize the paths of the Lasso. Report the top 10 selected genes. (Use the default variable names provided by R.) 3. Consider the SRBCT microarray data. (X and Y are in different files. Note the data does not provide column names, so when reading into R using read.csv, specify read.csv("X_SRBCT.csv", header=F".)) This is a gene expression data set from a childhood cancer study with n=83 patients and p=2308 genes. The response is the expression profile of p53, a major oncogene that acts as a tumor suppressor. The goal is to select a relatively smaller number of other genes whose expression profiles are associated with p53. a) Among the regression methods we learned, including least square, subset selection, ridge regression, principal component regression, and the lasso, which one(s) is (are) appropriate? Explain. b) Visualize the paths of the Lasso. Report the top 10 selected genes. (Use the default variable names provided by R.)