'Pends on what you want to happen w/ the rest of the data...it's trivial enough to simply augment a vector with zeros on either end but you might want to read
doc circshift
ADDENDUM
Well, if you have the two points of interest, then pick one as the reference; I'll presume to move the one with the rightmost peak and, for convenience in addressing the two that they're in a 2D array by column.
Given then nChMax is the corresponding 2-vector of the locations, to align the two would be
[~,idx]=max(nChMax); % determine which is the rightmost peak of the two
nZ=abs(diff(nChMax)); % the offset between the two in absolute mag
s(:,idx)=[s(nChMx(idx)-nZ:end,idx); zeros(nZ,1)]); % shift up by nZ and augment
This "throws away" the first of the shifted signal maintaining the same overall length. Alternatively, you can also augment the other with nZ at the front and keep the full length of signal but other than for visual purposes it won't matter as there's nothing matching when you get ready to do the calculations on a pointwise basis.
