I combined The CSI and NOE data by creating a Mutant factor and Residue factor. There were a few Mutants from NOE that I did not have CSI measurement for so they were excluded. Residues are positions 2:30.
I created factors for CSI and NOE as well. Based on the cutoffs we discussed. For the factor Structure I used CSI \(\geq\) 0 are considerd Loop, while -0.2 \(\leq\) CSI \(\leq\) 0 \(\Rightarrow\) No Structure and CSI \(\leq\) -0.2 \(\Rightarrow\) Helix. And for the factor Dynamics I used NOE \(\geq\) 0.6 as More and NOE \(\leq\) 0.6 as Less.
All of these cutoffs can be changes very easily. Here is a preview of what the data look like now.
## Mutant Residue CSI NOE Structure Dynamics
## 1 AFA 2 -0.14000 0.5558715 No Structure More
## 2 pS16_PLN 2 -0.06947 0.3769380 No Structure More
## 3 S16E 2 -0.14841 0.5061962 No Structure More
## 4 P21G 2 -0.14500 0.5874758 No Structure More
## 5 S16E_P21G 2 -0.07300 0.4986210 No Structure More
## 6 P21A 2 -0.11000 0.4802768 No Structure More
Some basic counts of the new factor variables for Structure and Dynamics
## `geom_smooth()` using method = 'loess'
## `geom_smooth()` using method = 'loess'
First the overall correlation between CSI and NOE
##
## Pearson's product-moment correlation
##
## data: data$CSI and data$NOE
## t = -11.233, df = 404, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.5586514 -0.4099949
## sample estimates:
## cor
## -0.4878522
And then the correlations by both Mutant and Residue