Introduction
Oneat identifies cell division events as spatiotemporal (TXYZ) coordinates. These coordinates were then used to impose trajectory splitting within TrackMate tracks through our custom-developed TrackMate-Oneat plugin. The plugin optionally applies the MARI (Mitosis Angular Region of Interest) principle, which filters division events to retain only those in which daughter cells emerge perpendicular to the mother cell’s nuclear major axis. Oneat integration significantly improved mitosis detection compared to TrackMate’s native linking algorithm.
By combining Oneat-predicted division locations with trajectory continuity, the system generated more biologically realistic branching structures and reduced the number of false positives typically produced by Oneat alone.
This biologically-informed relinking improves the completeness and accuracy of lineage tracking, especially in datasets with frequent cell divisions. To avoid spurious links and ensure geometric plausibility, the pipeline also incorporates the Mitosis Angular Region of Interest (MARI) principle. This constraint limits the search for daughter cells to a angular region from the mother cell’s nucleus principal axis of a fit ellipsoid. Candidate daughter spots were defined as those within a radial distance of the mother spot whose displacement vector , with the candidate spot position, formed an unsigned angle with the mother’s principal axis not exceeding a threshold set by the user. By restricting candidate daughters to fall within a defined angular region of interest, this method eliminates improbable pairings and enhances the biological realism of the reconstructed lineages. This constraint is especially important in dense tissues, where purely distance-based linking may result in incorrect associations.
Introduction
A TrackMate detector to track skeleton end points. It performs skeletonization of tissue endpoints, specifically, it creates a SpotCollection by skeletonizing an integer or binary segmented image, enabling the tracking of growing tissue branches github over time. It takes a segmented microscopy image, reduces the tissue structures to their skeletal form, detects the branch endpoints as trackable spots, and feeds them into TrackMate’s tracking pipeline, making it particularly useful in developmental biology workflows for quantifying how tissue branches grow and change across frames.
