LecturesPart25

LecturesPart25 - Computational Biology Part 25 Image-based...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Computational Biology, Part 25 Image-based Cell Models Robert F. Murphy Copyright © 2005-2009. All rights reserved. Generative models of subcellular patterns LAMP2 pattern Cell membrane Nucleus Protein Nuclear Shape - Medial Axis Model width Rotate Represented by two curves Medial axis the medial axis width along the medial axis Synthetic Nuclear Shapes With added nuclear texture Cell Shape Description: Distance Ratio d1 d2 d1 + d 2 r = d2 Capture variation as a principal components model Generation Modeling Vesicular Organelles Original Filtered Fitted Gaussians Object Positions d1 d2 d2 r = d1 + d 2 Models for protein-containing objects s r: normalized distance, a: angle to major axis Mixture of Gaussian objects s Learn distributions for number of objects and object size s Learn probability density function for objects relative to nucleus and cell Synthesized Images Lysosomes Endosome s SLML toolbox - Ivan Cao-Berg, Tao model parameters Have XML design for capturing Peng, Ting Zhao 12 Have portable tool for generating images from model Model Distribution s s s Generative models provide better way of distributing what is known about “subcellular location families” (or other imaging results, such as illustrating change due to drug addition) Have initial XML design for capturing the models for distribution Have portable tool for generating images from the model Generation Process Protein Cell Shape Nuclear Model XML Generating Multiple Distributions for Simulations Simulation 1 Protein Cell Shape Nuclear Model XML Simulation 2 Conclusions Simulation 3 Combining Models for Cell Simulations Protein 1 Cell Shape Nuclear Model Protein 2 Cell Shape Nuclear Model Protein 3 Shared Nuclear and Cell Shape Cell Shape Nuclear Model XML Simulation Example combination Red = nuclear membrane, plasma membrane Blue = Golgi Green = Lysosomes Direct vs Indirect Generative Modeling s Direct Method: x Estimate model parameters directly from images x Hard for microtubule networks s Indirect (Inverse) Method: x Compare images with synthetic images generated from model to find parameters x Example: work of David Odde group on spindle length distribution Final word s Goal of automated image interpretation should not be x Quantitating intensity or colocalization x Making it easier for biologists to see what’s happening s Goal should be generalizable, verifiable, mechanistic models of cell organization and behavior automatically derived from images Biochemical models that consider compartment geometry s Virtual Cell facilitated Ca-diffusion model from tutorial s http://www.nrcam.uchc.edu/login/facil_ca_ dif.pdf Making a compartment map for Virtual Cell from a fluorescence microscope image Start from a fluorescence microscope image of a lysosomal protein (LAMP-2) Making a compartment map for Virtual Cell from a fluorescence microscope image s Use Matlab to create an image with values of zero for background, one for cytoplasm, and two for lysosomes s Assume that the autofluorescence in the lysosome image is sufficient to find a region corresponding to the cytoplasm Make contiguous cytoplasm image by averaging weak autofluorescence img=imread('r06aug97.h4b4.13--1---2.dat.png'); a=double(img); b=(a-min(min(a)))./(max(max(a))-min(min(a))); H=fspecial('average',13); c=imfilter(b,H,'replicate'); d=im2bw(c,0.004); imshow(d); max(max(d)) Combine with image of pixels with positive lysosomal staining e=im2bw(b,graythresh(b)); imshow(e); f=d + e; imshow(f,[0 2]); g=uint8(f); imwrite(g,'geomap.tif','TIF','Compression','none'); Resulting image Virtual Cell-PSLID interface ...
View Full Document

This note was uploaded on 12/03/2011 for the course BIO 118 taught by Professor Staff during the Fall '08 term at Rutgers.

Ask a homework question - tutors are online