The Open Microscopy Environment Analysis System: Workflow Composition and Enactment for Quantitative Pattern Analysis of Large Microscopy Datasets

Add to your list(s) Download to your calendar using vCal

• Tom Macura, Computer Laboratory, University of Cambridge
• Tuesday 05 February 2008, 14:30-15:30
• Room FW11, Computer Laboratory, William Gates Building.

If you have a question about this talk, please contact Minor Gordon.

The Open Microscopy Environment (OME) Analysis System is software that supports modelling and enacting workflows for the quantitative analysis of microscopy images. This system is based on OME – an extensible infrastructure with plug-in ontologies for managing large sets of biological images and data. Integrated into the analysis system is a multi-purpose image classifier for scoring high content screens (HCS) and other high throughput imaging applications. The image classification workflow is composed of 53 nodes with 189 links that output 1025 numerical values modeled as 48 ontological terms. Nodes are MATLAB scripts, with unaltered source code, around which XML execution instructions wrappers have been written to incorporate the scripts into OME . Users can convert their legacy image analysis tools into OME workflows or customize the integrated image classification workflow to suit their task-specific needs.

The OME Analysis System has been validated on an example high content, high throughput experiment called the Atlas of Gene Expression in Mouse Aging Project (AGEMAP). The AGEMAP dataset is ~30,000 microscope images of mouse livers, skeletal muscles, and kidney tubules. Beginning with 18GB of raw pixels, it took the OME approximately 2000 processor hours (luckily, OME supports distributed computation) to generate 125GB of intermediary pixels and extract 90 million image descriptor features used for classification.

In this presentation I will briefly introduce high content, high throughput biological imaging, describe the design and development of OME as well its validation (including performance) on AGEMAP , and conclude by mentioning some of the biological insights we learned about aging from the AGEMAP analysis.

This talk is part of the Computer Laboratory Opera Group Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge talks
• Computer Laboratory Opera Group Seminars
• Department of Computer Science and Technology talks and seminars
• Interested Talks
• Room FW11, Computer Laboratory, William Gates Building
• School of Technology
• Trust & Technology Initiative - interesting events
• bld31

Note that ex-directory lists are not shown.