lab10_rnaseqIII_280

## No vignettes or demos or help files found with alias or concept or ## title matching 'cor_matrix' using regular expression matching. Retrieve correlation cor_matrix[ 1 , 2 ] ## [1] 0.6935682 Q4. Notice that the values along the diagonal of the matrix are all 1’s. What is the reason for this? Answer: The values along the diagonal of the matrix are all 1’s because they represent the pairwise correlation of a variable with itself. When you compare a variable with itself it will be perfectly correlated because it is identical and as a result this perfect correlation will result in a score of 1. Hierarchical Clustering hc <- hclust ( as.dist ( 1 - cor_matrix)) Plot Hierarchical Clustering plot (hc)

Q8. Why is the correlation heatmap symmetric? Answer: Similar to why we see the diagonal in the correlation matrix for pairwise correlations, we see symmetry in the heat map because we are comparing the same variables and their correlations in symmetric regions of the heatmap so we will get exact correlation matches in those symmetric regions. Calculate variance of all row var (raw_counts) ## procyclic_1 metacyclic_1 procyclic_2 metacyclic_2 procyclic_3 ## procyclic_1 191039681 69137405 162836927 60149036 206552862 ## metacyclic_1 69137405 52014470 72698074 47855254 98614052 ## procyclic_2 162836927 72698074 156597600 61258748 204872111 ## metacyclic_2 60149036 47855254 61258748 51447263 81708007 ## procyclic_3 206552862 98614052 204872111 81708007 279409143 ## metacyclic_3 66809947 52788386 69770623 56211143 96247654

Q11. In the PCA plot created above, which sample appears to differ the most from the other samples in its condition? Answer: procyclic_3 because it appears to be far away from other procyclic samples (the other samples seem to be clustered toghther). Calculated Total Variance round (fit $ sdev ** 2 / sum (fit $ sdev ** 2 ), 2 ) * 100 ## [1] 48 23 8 7 4 4 3 2 1 0 PCA Plot with Variance Explained var_explained <- round (fit $ sdev ** 2 / sum (fit $ sdev ** 2 ), 2 ) * 100 plot (fit $ x[, 1 ], fit $ x[, 2 ], col= rep ( c ( 'red' , 'blue' ), 5 ), xlab= sprintf ( "PC1 (%d%% variance)" , var_explained[ 1 ]), ylab= sprintf ( "PC2 (%d%% variance)" , var_explained[ 2 ])) text (fit $ x[, 1 ], fit $ x[, 2 ], colnames (raw_counts_filtered), pos= 3 ) Q12. Which aspect of the samples does PC1 appear to correspond to in the plot above?