Alphan Altinok
Research Associate, Center for Bioimage Informatics, Dept of Electrical and Computer Engineering, University of California, Santa Barbara.
22 Tuesday, Sept 2009, 12:30PM - 1:30PM
Steele 214
Biological image data are at least as diverse as the experimentation spectrum itself. Even working with a unique cell component, computational tools require frequent adaptation to various types of data and analysis. In this talk, I will present our work on quantitatively analyzing dynamic properties of a particular cell component: microtubules. I will highlight different analysis requirements on the same imaging modality, as well as different imaging modalities and associated divergence in quantification and analysis. To ensure effective use, computational methods should provide intuitive interfaces for general use, and in some cases for user interaction. I will provide examples of handling user interaction.
Microtubules are cytoskeletal polymers that participate in several critical cell functions. Their regulation mechanisms have long been implicated in developmental disorders and several devastating diseases, including Alzheimer's Disease and cancer. Therefore, the properties of their dynamic instability are of significant interest, yet the traditional image analysis methods remain arduous. By automating the procedures of quantification and analysis, we aim to address this challenge. However, as is evident from the diversity in imaging and experimental setups, frequent algorithmic and user-interactivity adjustments are necessary. Before converging on universally applicable methods of biological image analysis, we may need to utilize ad-hoc solutions for pragmatic reasons. This is also motivated by the evolving landscape in biological experimentation as well as in imaging methods. I will illustrate this case on our recent work, where microtubule dynamics are observed only indirectly.
