Loading…
Describing Shape Dynamics in Transformed Cells through Latent Factors
Previous studies attributed the characteristic shape changes found in cancer cells, in part, to aberrant vesicle traffic. Typically, transformed cells also rounded up. These phenomena were further investigated by measuring the shape features of cells from established lines, which represented both no...
Saved in:
Published in: | Experimental cell research 1999-01, Vol.246 (1), p.69-82 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Previous studies attributed the characteristic shape changes found in cancer cells, in part, to aberrant vesicle traffic. Typically, transformed cells also rounded up. These phenomena were further investigated by measuring the shape features of cells from established lines, which represented both normal and oncogenic stages of transformation. Although conventional pattern recognition methods, applied to a combined data set from these lines, failed to reveal any new, recognizable features beyond those already known, factors did describe such features. Factors are hypothetical variables that contribute to the variance of two or more measurable variables. One factor for the cell edge, 5, was known from previous studies on correlations among the variables. Several other factors at the same level identified crucial features. Factor 4 reflected the frequency of microspikes; another factor described a knob-like structure (7). A third, factor 16, indexed the variability in projection size. Factors of the upper cell, 1 μm or more above the substratum, namely, 1, 2, 8, 11, 13, and 19, also described transformation-related changes. Comparing lines that modeled the development of bronchogenic carcinoma, we found a tendency for 2 (surface smoothing), 4, and 12 (rounding-up) to be changed irreversibly. Thus, factors overcame the problem of relating mathematical shape phenotypes, previously obtained based on single variables, to cell features. |
---|---|
ISSN: | 0014-4827 1090-2422 |
DOI: | 10.1006/excr.1998.4242 |