Construct a data frame to test a new classification algorithm when the null hypothesis is true. The data frame should contain "microarray data" from 5 genes monitored across 15 patients. Since this is the null hypothesis, the numbers should be all random. Construct a data frame containing 50 numbers from a random normal distribution. Label the first 5 samples with "cancer," the following 5 with "treated," and the remaining 5 samples with "healthy." Calculate the mean of each gene in each group. Plot the distribution of the gene values using a different color for each group.