Quantification of cellular contributions to multi-cellular pattern formation

Life is built on patterns. From the stripes on a zebra to the arrangement of leaves on a stem, living things constantly organize themselves into complex, repeating structures. This process, known as pattern formation, is fundamental to how an animal grows from a single cell into a fully formed organism, how it heals after an injury, and unfortunately, how diseases like cancer spread. For decades, scientists have tried to understand the engine behind this organization. They have long known that cells can change their identity, move around, divide, or die, and that any of these actions could potentially create the large-scale designs we see in nature. The challenge has been figuring out exactly which of these cellular actions is the primary driver in any given situation. Is a pattern created because cells are moving to specific spots, or because they are dividing more in some areas than others? Until now, answering this question usually required researchers to guess a specific mechanism first, build a complex mathematical model based on that guess, and then see if the model matched reality. If the guess was wrong, the whole investigation had to start over.

A new approach, developed by researchers at Aix Marseille University and EMBL Heidelberg, bypasses the guessing game entirely. Instead of starting with a theory about how cells should behave, they created a method to simply watch what the cells are doing and measure how much each specific action contributes to the final pattern. They call their tool CREDIT, which stands for Cellular Roles in Emergent Development Inferred from Tracks. The method relies on high-quality video footage of living cells, where scientists can track the exact position of every single cell over time and measure how much of a specific protein each cell is producing. By analyzing these tracks, the researchers can mathematically separate the total pattern into its component parts, quantifying exactly how much of the design is due to cells moving, how much is due to cells dividing, and how much is due to cells changing their internal chemistry.

The researchers first tested their method using computer simulations where they knew the answer beforehand. They created virtual worlds where they could turn the "volume" of cell movement up or down, and do the same for cell division and death. In these controlled simulations, the tool correctly identified the dominant force. When they told the virtual cells to move to form a pattern, the tool said movement was the cause. When they told the cells to divide in specific spots to create the design, the tool identified division as the driver. This validation proved that the method works without needing to know the underlying rules of the system; it simply observes the geometry of the cells and the resulting pattern to deduce the cause.

With the tool proven reliable, the team applied it to real biological data. The first subject was a cluster of mouse embryonic cells, specifically from a tissue called the presomitic mesoderm, which is crucial for forming the backbone and muscles. In these experiments, the cells formed distinct clusters and patterns of a protein called T/Brachyury. Previous studies had suggested that the cells might be moving to these spots because of chemical signals, or that the cells were changing their protein levels to create the pattern. However, when the researchers ran their new analysis, the result was clear and decisive: the pattern was almost entirely driven by the physical motion of the cells. The cells moved around and aggregated into clusters, and this movement alone was sufficient to create the observed pattern. The changes in the cells' internal protein levels played a negligible role in shaping the design. This finding supports the idea that mechanical forces and cell movement are the primary architects of this specific developmental stage.

The researchers then turned their attention to a very different system: breast cancer cell cultures. In these dishes, cancer cells organize themselves into "niches," or clusters, that resemble the environments where stem cells usually live. These clusters are marked by a specific protein, ALDH1A1, which is associated with cancer stem cells. Understanding how these clusters form is vital because they are often responsible for tumors resisting treatment. In this case, the new method revealed a completely different story. Here, the pattern was not driven by cells moving to specific locations. Instead, the analysis showed that the formation of these cancer stem cell niches was driven almost exclusively by cell division. The cells were dividing more frequently in certain areas, creating dense clusters. Interestingly, the researchers found that the changes in the cells' protein levels actually worked against the formation of the pattern, slightly disrupting the design, but this negative effect was overwhelmed by the sheer force of cell division.

The power of this new approach lies in its simplicity and its lack of bias. It does not require scientists to assume a specific mechanism is at work before they begin. It treats the cells as geometric objects moving in space and asks a simple question: what is the cell doing right now that changes the overall picture? By breaking down the complex dance of development into distinct, measurable contributions, the method reveals that the answer is not always the same. In the mouse embryos, movement is the key; in the cancer cells, division is the key. This distinction is critical because it tells biologists exactly where to look for the next step in their research. If a pattern is driven by movement, researchers should study the mechanical forces between cells. If it is driven by division, they should look at the signals that control the cell cycle.

This work represents a shift from hypothesis-driven science to data-driven discovery. For years, the field has been stuck trying to fit observations into pre-existing models. This new tool allows the data to speak for itself, quantifying the contribution of every cellular process without needing to know the rules of the game in advance. It provides a clear, unbiased map of how life organizes itself, showing that the path to a pattern can be paved by movement, by division, or by a combination of both, depending entirely on the specific biological context. As imaging technology continues to improve, allowing scientists to track cells in three dimensions and in real-time, this kind of geometric analysis will likely become a standard way to understand the hidden mechanics of life, from the earliest stages of an embryo to the progression of disease.