Cell Morphology accurately predicts the nuclear shape of adherent cells
Lawton, S.; Danzman, R. A.; Spagnuolo, R.; Stephan, S.; Graul, S.; Wiggan, O.; Ghosh, S.; Prasad, A.
Show abstract
Cells are internally tensed, or prestressed, largely by actomyosin contractility. We hypothesized that nuclear shape is quantitatively predictable from cell shape since prestress couples them both. We trained machine learning models on a publicly available image database of the WTC-11 cell line and predicted shape modes of the nucleus with high accuracy. We develop a U-Net architecture-based model, Cell2Nuc, that predicted nuclear voxels from the cell membrane with accuracies between 74%-87%. To investigate prestress, we cultured and imaged HeLa cells after inhibiting actomyosin contractility. The Cell2Nuc model retrained on the HeLa cells predicted nuclear voxels with slightly lower accuracy. Statistical analysis revealed changes in nuclear size and chromatin organization upon prestress inhibition. Similar trends were seen in images taken from NIH3T3 cells. Thus, cell shape encodes features of nuclear shape, their coupling is partly due to actomyosin contractility, whose abrogation leads to changes in chromatin organization of mechanosensitive origin.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Shape-to-graph Mapping Method for Efficient Characterization and Classification of Complex Geometries in Biological Images 95%
- A deep generative model of 3D single-cell organization 95%
- Interkinetic nuclear movements promote apical expansion in pseudostratified epithelia at the expense of apicobasal elongation 94%
Similar papers in this journal
- High-volume, label-free imaging for quantifying single-cell dynamics in induced pluripotent stem cell colonies 95%
- A comprehensive suite of measures to quantify cell shape and actin organization 94%
- Using a continuum model to decipher the mechanics of embryonic tissue spreading from time-lapse image sequences: An approximate Bayesian computation approach 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.