deletions | additions       diff --git a/In_the_case_of_CA__.tex b/In_the_case_of_CA__.tex  index cd4d9f5..f452b0a 100644  --- a/In_the_case_of_CA__.tex  +++ b/In_the_case_of_CA__.tex 
...
In the case of CA C$\alpha$  none of the terms have a large effect on the chemical shielding. In the case of 2OED GB3  the results improve slightly if $\Delta \sigma_{1^\circ HB}^i$ is removed and removing $\Delta \sigma_{1^\circ H\alpha B}^i$ improves the results slightly for both proteins. We found that the affects are not always additive (presumably due to error cancellation) and that the overall RMSD could be lowered further (to 1.6 ppm) by also removing $\Delta \sigma_{2^\circ HB}^i$. In the case of $\Delta \sigma_{1^\circ HB}^i$ and $\Delta \sigma_{2^\circ HB}^i$ this is not particularly surprising as CA C$\alpha$  is approximated as a methyl group in the structural model. Accordingly these three terms are removed from ProCS15, while all other terms are kept (note the ring current is only included for hydrogen atoms). For CB C$\beta$  removing $\Delta \sigma_{1^\circ H\alpha B}^i$ decreases the RMSD by 0.2 - 0.5 ppm, while $\Delta \sigma_{BB}^{i-1}$ and $\Delta \sigma_{BB}^{i+1}$ increases and decreases the RMSD value depending on the protein. Accordingly only $\Delta \sigma_{1^\circ H\alpha B}^i$ is removed. Note that the structural models used for $\Delta \sigma_{1^\circ HB}^i$ and $\Delta \sigma_{2^\circ HB}^i$ do not contain a CB C$\beta$  so there is no such contribution for this nucleus. For C C'  removing $\Delta \sigma_{1^\circ H B}^i$ decreases the RMSD for 2OED GB3  by 0.1 so we choose to remove this term for this atom type. Note that removing $\Delta \sigma_{2^\circ H\alpha B}^i$ increases the RMSD by 0.4 - 0.6 ppm so this term is important for accurate predictions of C C'  chemical shifts. For H and HA H$\alpha$  we choose to retain all the terms. Not surprisingly, the respective primary hydrogen bonding terms lower the RMSD by 0.4 - 0.6 ppm and are crucial for accurate predictions. Finally, for N removing the hydrogen bonding terms involving HA tends to lowers the RMSD slightly, so these two terms are removed. Note that $\Delta \sigma_{BB}^{i-1}$ and the two hydrogen bonding terms involving H lower the RMSD by as much as 2.6 ppm ($\Delta \sigma_{BB}^{i-1}$ for 1UBQ) ubiquitin)  and is crucial for accurate predictions.