Average trajectories can be aligned in space and in time together by chosing one protein, and its average trajectory, as a spatial and temporal landmark. For simplicity, we call this protein reference protein and its trajectory reference trajectory, while the protein that needs to be alinged is referred to as target proteins and its dynamics are described by the target trajectory.

The functions necessary to average the different trajectories together are defined in To align the target trajectory to the reference trajectory we need to import the function align in align.py and the function load_directories in average.py:

from trajalign.align import align
from trajalign.average import load_directory

load_directory is used to load all the trajectories that are acquired simultaneously for both the target and the reference proteins in two distinct lists. These trajectories will be used to compute the transformation that alinged the target and reference trajectories together. It is important that the trajectories in these lists are ordered in the same way, so that the first trajectory of the reference protein in one list and the first trajectory of the target protein in the other list have been acquired in the same event. Also, it is important the
relative position between the trajectories of the reference and target proteins have been corrected for chromatic aberrations. See the example for details on how to load the trajectories that are used to compute the alignment of the reference and target average trajectories.

align( path_target , path_reference , ch1 , ch2 , fimax1 = False , fimax2 = False ) align the target average trajectory, identified by the path path_target to the reference average trajectory, identified by the path path_reference. ch1 and ch2 are the lists of the trajectories for the reference and target protein that have been acquired simultaneosly and which are used to compute the alignment. fimax1 and fimax2 are logic values that instruct align whether to use only the trajectory information up to the peak in fluorescence intensity (True) or to use the complete trajectory information (False, which is the default). It is important when trajectories are particularly noisy and dynamic toward the end of their lifetime, as it is the case for Abp1 in endocytosis.

Aligned trajectories are saved with the same name as the target trajectory (path_target), followed by “aligned”. These trajectories store also new annotations: the name of the target trajectory, the name of the reference trajectory and the transformation and its error that aligns the target trajectory to the reference trajectory. The transformation is computed as the median and the errors are the standard errors computed from the MAD adjusted for asymptotically normal consistency (k = 1.4826).