Finding the dynamic structure of kinetochores using deep learning reconstruction

Susan Cox (primary)
Randall Centre
King's College London
Philip Auckland (secondary)
Randall Centre
King's College London

Abstract

Mitosis is essential for the growth, development and homeostasis of eukaryotes. Kinetochores are protein machines that control mitotic progression by forming dynamic force-generating attachments to spindle microtubules. Here, we will use super-resolution microscopy to image large numbers of kinetochores at distinct mitotic stages at a resolution of tens of nm. A cutting-edge deep learning approach will be used to fit the data, where the training process of the network is used to optimise a model with no prior constraints being applied. This model will be integrated from data from live-cell experiments to create the first nanoscale model of kinetochore dynamics.


References

1 Auckland, Pᵼ., Roscioli, R., Coker, H.L.E., and McAinsh, A.Dᵼ. (2020). CENP-F stabilizes kinetochore-microtubule attachments and limits dynein stripping of corona cargoes. J Cell Biol, doi: 10.1083/jcb.201905018 ᵼco-correspondence

2 Auckland, P., Clarke, N., Royle, S. & McAinsh, A.D. (2017). Congressing kinetochores progressively load Ska complexes to prevent force dependent detachment. J Cell Biol, doi: 10.1083/jcb.201607096

3 Marsh RJ, Pfisterer K, Bennett P, Hirvonen LM, Gautel M, Jones GE, Cox S*, ‘Artifact-free high-density localization microscopy analysis’, 2018, Nature Methods, 15:689

4 Mahdi Rad, Vincent Lepetit, BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects Without Using Depth, Proceedings of the ICCV, 2017, 3828-3836


BBSRC Area
Molecules, cells and industrial biotechnology
Area of Biology
Cell BiologyStructural Biology
Techniques & Approaches
BiochemistryBiophysicsImage ProcessingMicroscopy / ElectrophysiologyMolecular Biology