Recent advances in imaging technology allow researchers to capture snapshots of embryonic cells as they divide and migrate through development. But capturing an image of a growing embryo every 30 seconds or so, produces terabytes of data that is time consuming to analyze. A team led by HHMI, Janelia Group Leader, Philipp Keller, Ph.D., has created a piece of software that identifies and tracks dividing cells as quickly as high-speed microscopes can collect the images. By clustering pixels and using these
clusters as the smallest unit of measurement, the researchers reduced complexity by a thousand-fold. The computer scans for connected groups of these clusters for shapes resembling nuclei and that information is then used to locate those nuclei in subsequent images.
Keller’s lab, as reported on July 20, 2014, in Nature Methods, was able to track a large fraction of early neuroblasts and could even predict the future fate and function of many cells based on their behavior.
Because Keller hopes others will use the program, which works with data from several different types of fluorescent microscopes, to advance the field of early development research, he and his team have made the software freely available on his website.