Today, tracking the development of individual cells and spotting the particular associated factors under the microscope is nothing unusual. Nevertheless , impairments like shadows or changes in the background complicate the particular interpretation of data. Now, researchers at the Technical University or college of Munich (TUM) and the Helmholtz Zentrum Mü nchen have developed a software that corrects images to make hitherto concealed development steps visible.

Whenever stem cells develop into specialized cells, this happens in several steps. But which regulatory proteins are active throughout the decisive branching on the development path? Using so-called time-lapse microscopy, researchers can observe individual cells at quite high time resolutions and, using fluorescent labelling, they can identify precisely which of these proteins appear when in the cellular.

Once a stem cell has been identified, it could be closely observed over several days using cell-tracking software program. Yet, this “surveillance work” often turns out to be difficult. “The imaging data is frequently marred by irregular brightness plus faded backgrounds in the time-lapse, ” explains Dr . Carsten Marr, heading the workgroup Quantitative Single Cell Characteristics at the Institute of Computational Biology (ICB) of the Helmholtz Zentrum Mü nchen. “This makes it difficult or unattainable to detect proteins that are decisive when a cell opts for a specific development direction, so-called transcription factors. inch

Algorithms that filter out these kinds of artefacts can be found, but they require either specifically prepared reference images, numerous images per dataset or complex manual adjustments. In addition, none of the existing methods correct alterations in the background as time passes, which hamper the quantification of individual cells.

Algorithm eliminates background changes

Now, Dr . Tingying Peng, member of Dr . Carsten Marr’s group at the Helmholtz Zentrum Mü nchen plus Professor Nassir Navab, head of the Chair for Personal computer Aided Medical Procedures and Augmented Reality at TU Munich, present an algorithm that corrects these artefacts using only a couple of images per dataset.

The software is called “BaSiC” and is freely available. It is compatible with many image forms commonly used in bioimaging, including mosaics pieced together through numerous smaller images and used, for example , to provide large tissue regions. “Contrary to other programs, however , inch explains Dr . Peng, “BaSiC can correct changes in the history of time-lapse videos. This makes it a valuable tool intended for stem cell researchers who want to detect the appearance of particular transcription factors early on. ”

Getting significant details to light

Exactly how well the new image correction program improves the evaluation of individual stem cell development steps the researchers demonstrated with time-lapse videos of blood stem tissue. They recorded the videos to observe cells over a six-day time span. At a certain point during this observation period undifferentiated precursor cells choose between two possible tacks of growth that lead to the formation of different mature blood tissue.

In images corrected using BaSiC, the particular researchers could identify a substantial increase in the intensity of a particular transcription factor in one of the two cell lines, while the quantity of his protein in the other cell line remained unrevised. Without the image correction, the difference was not ascertainable.

“Using BaSiC, we were able to make important decision elements visible that would otherwise have been drowned out by sound, ” says Nassir Navab. “The long-term goal of the research is to facilitate influencing the development of stem cells in the targeted manner, for example to cultivate new heart muscle tissue cells for heat-attack patients. The novel possibilities meant for observation are bringing us a step closer to this objective. ”