|
Activity Number:
|
225
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2004 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #300598 |
|
Title:
|
Segment of Plasma Membrane Image and Spatial Statistical Analysis of PM Sterol Distribution
|
|
Author(s):
|
Weimin Zhang*+ and Avery L. McIntosh and Barbara Atshaves and Friedhelm Schroeder and Jyh-Charn Liu and Suojin Wang
|
|
Companies:
|
Texas A&M University and Texas A&M University and Texas A&M University and Texas A&M University and Texas A&M University and Texas A&M University
|
|
Address:
|
Dept. of Statistics, College Station , TX, 77843,
|
|
Keywords:
|
segmentation ; rank statistic ; union intersection test ; completely spatial randomness test ; Monte Carlo simulation
|
|
Abstract:
|
Although plasma membrane (PM) cholesterol-rich and -poor domains have been isolated by subcellular fractionation, it has been unclear whether such domains exist in real-time in living cells. Therefore, dehydroergosterol (DHE), a naturally occurring fluorescent sterol, is incorporated into cultured L-cell fibroblast and visualized in real-time in PM of living cells by multiphoton laser scanning microscopy (MPLSM). Rank statistics are used for segmenting PM area from MPLSM image of living cells. The Miller's rank statistic is calculated for combined two areas and the union intersection test is used for multivariate (color) data comparisons. Spatial distributions of DHE within the PM are examined by the complete spatial randomness (CSR) test and a polynomial regression analysis is run for prediction intervals. A Monte Carlo simulation study shows that the distributions of DHE appear in spatial clusters. These results for the first time demonstrate that sterol distributes clusterly in the membranes of living cells.
|
