[Crossposted at A Man with a PhD]
4-dimensional live cell imaging has gone from being a rare technique used only by cutting-edge laboratories to a mainstream method in use everywhere. While more and more labs are becoming comfortable with the equipment and protocols needed to collect imaging data, performing detailed analyses is often problematic. The application of computational image processing is still far from routine. Researchers need to determine which measurements are necessary and sufficient to characterize a system and they need to find the appropriate tools to extract these data. In Computational Image Analysis of Cellular Dynamics: A Case Study Based on Particle Tracking, Gaudenz Danuser and Khuloud Jaqaman introduce the basic concepts that make the application of computational image processing to live cell imaging data successful. As one of the featured articles in December’s issue of Cold Spring Harbor Protocols, it is freely accessible for subscribers and non-subscribers alike.
The article is adapted from the new edition of Live Cell Imaging: A Laboratory Manual, now available from CSHL Press.
My first year as a biochemistry graduate student, one of the classes simply dealt with the analytical technologies we would be using. Things like NMR, UV spectroscopy, circular dichroism, fluorescence and X-ray crystallography. They would help us understand the properties of isolated biological molecules
This paper gives a great view of some of the new analytical approaches that examine entire living cells, not just isolated molecules. Now it looks like students will also have to get some firm understanding of image analysis. There will be some really interesting results from these sorts of technologies. The conclusions provide insights into the promise and the problems:
Computational image analysis is a complex yet increasingly central component of live cell imaging experiments. Much has to be done to make these techniques useful for cell biological investigation. First, algorithms must be transparent, not necessarily at the level of the code but in terms of their sensitivity to changing image quality and the effect that control parameters have on the output. Second, the design of imaging experiments must be tightly coupled to the design of the analysis software. All too often, images are taken without careful consideration of the subsequent analysis and are forwarded to the computer scientist to retrieve information from the images. To avoid these problems, communication must be initiated early on, and experiments must be designed with the appreciation that data acquisition and analysis are equivalent components. Third, software development and application require careful controls, as is customary for molecular cell biology experiments. This article provides a brief introduction to the ideas useful for implementing such controls. Hopefully, the cell biological literature will include a more extensive discussion of the measures taken to substantiate the validity of results from image analysis. On the other hand, manual image analysis should no longer be an option. As discussed in this article, manual analyses fall short in consistency and completeness, two essential criteria underlying the validity of a scientific model derived from image data.
While the results can be amazing, there needs to be close collaboration between the different researchers involved. Because very few people will have all the expertise necessary for success. This tight coupling of researchers with vastly different backgrounds and focus (i.e. cell biology and bioinformatics) is a relative new aspect of modern biological research.
There may be slowing of this coupling in some labs but the successful results by those that can accomplish this type of collaboration will rapidly overtake those who take a slower course. As I mentioned below, large collaborations may be a big part of the published record as we move forward.