Deciphering anomalous heterogeneous intracellular transport with neural networks
- University of Manchester, United Kingdom
Abstract
Intracellular transport is predominantly heterogeneous in both time and space, exhibiting varying non-Brownian behavior. Characterization of this movement through averaging methods over an ensemble of trajectories or over the course of a single trajectory often fails to capture this heterogeneity. Here, we developed a deep learning feedforward neural network trained on fractional Brownian motion, providing a novel, accurate and efficient method for resolving heterogeneous behavior of intracellular transport in space and time. The neural network requires significantly fewer data points compared to established methods. This enables robust estimation of Hurst exponents for very short time series data, making possible direct, dynamic segmentation and analysis of experimental tracks of rapidly moving cellular structures such as endosomes and lysosomes. By using this analysis, fractional Brownian motion with a stochastic Hurst exponent was used to interpret, for the first time, anomalous intracellular dynamics, revealing unexpected differences in behavior between closely related endocytic organelles.
Data availability
Supporting files are on GitHub and Zenodo.
Article and author information
Author details
Funding
Wellcome Trust (215189/Z/19/Z)
- Daniel Han
EPSRC (EP/J019526/1)
- Nickolay Korabel
- Victoria J Allan
- Sergei Fedotov
- Thomas A Waigh
BBSRC (BB/H017828/1)
- Victoria J Allan
Wellcome Trust (108867/Z/15/Z)
- Anna Gavrilova
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Copyright
© 2020, Han et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 1,904
- views
-
- 306
- downloads
-
- 44
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Deciphering anomalous heterogeneous intracellular transport with neural networks
eLife 9:e52224.
https://doi.org/10.7554/eLife.52224
Further reading
-
- Cell Biology
Meiotic progression requires coordinated assembly and disassembly of protein complexes involved in chromosome synapsis and meiotic recombination. Mouse TRIP13 and its ortholog Pch2 are instrumental in remodeling HORMA domain proteins. HORMAD proteins are associated with unsynapsed chromosome axes but depleted from the synaptonemal complex (SC) of synapsed homologs. Here we report that TRIP13 localizes to the synapsed SC in early pachytene spermatocytes and to telomeres throughout meiotic prophase I. Loss of TRIP13 leads to meiotic arrest and thus sterility in both sexes. Trip13-null meiocytes exhibit abnormal persistence of HORMAD1 and HOMRAD2 on synapsed SC and chromosome asynapsis that preferentially affects XY and centromeric ends. These major phenotypes are consistent with reported phenotypes of Trip13 hypomorph alleles. Trip13 heterozygous mice exhibit meiotic defects that are less severe than the Trip13-null mice, showing that TRIP13 is a dosage-sensitive regulator of meiosis. Localization of TRIP13 to the synapsed SC is independent of SC axial element proteins such as REC8 and SYCP2/SYCP3. Terminal FLAG-tagged TRIP13 proteins are functional and recapitulate the localization of native TRIP13 to SC and telomeres. Therefore, the evolutionarily conserved localization of TRIP13/Pch2 to the synapsed chromosomes provides an explanation for dissociation of HORMA domain proteins upon synapsis in diverse organisms.
-
- Cell Biology
Immunofluorescence localises proteins via fluorophore-labelled antibodies. However, some proteins evade detection due to antibody-accessibility issues or because they are naturally low abundant or antigen density is reduced by the imaging method. Here, we show that the fusion of the target protein to the biotin ligase TurboID and subsequent detection of biotinylation by fluorescent streptavidin offers an ‘all in one’ solution to these restrictions. For all proteins tested, the streptavidin signal was significantly stronger than an antibody signal, markedly improving the sensitivity of expansion microscopy and correlative light and electron microscopy. Importantly, proteins within phase-separated regions, such as the central channel of the nuclear pores, the nucleolus, or RNA granules, were readily detected with streptavidin, while most antibodies failed. When TurboID is used in tandem with an HA epitope tag, co-probing with streptavidin and anti-HA can map antibody-accessibility and we created such a map for the trypanosome nuclear pore. Lastly, we show that streptavidin imaging resolves dynamic, temporally, and spatially distinct sub-complexes and, in specific cases, reveals a history of dynamic protein interaction. In conclusion, streptavidin imaging has major advantages for the detection of lowly abundant or inaccessible proteins and in addition, provides information on protein interactions and biophysical environment.
