Extended Data Fig. 2: Schematic representation of the DTC image J plugin and the quantification pipeline of dynamic interaction.

(a) The plugin first detects each individual compartments in both channels for all time points and calculate their centroids. (b) The tracking of individual compartments is performed one step at a time by following the iterative presence of a compartment at t_n within its 412 nm (9 pixels) radius zone at t_n+1. (c) An association is detected if a compartment of one channel is crossing the association-zone of a compartment of the opposite channel. Oppositely, a compartment of one channel that does not cross any association zone of the opposite channel is independent. (d) The association zone is defined for each couple of markers by calculating the mean distance between two compartments of opposite channels (each centre is labelled by a blue cross) when they were the closest possible without being co-localized. (e) Each compartment is followed for a certain time before it disappears from the field of acquisition, this defines the track time. When a compartment of a channel is crossing the association zone (light purple) of a compartment of the opposite channel, an association event is detected. This association can last for a certain amount of time before it leaves the association zone (dissociation event), this defines the association time. Over a track time, several association/dissociation events can occur. (f–h) For a given track, three behaviours were classified, either a compartment of one channel remains associated to another of the opposite channel during the whole time of the track (f), remains independent during the whole time of the track (g) or show transient interaction possibly involving several association/dissociation events (h).