A novel method for analysis of the periodicity of chondrogenic patterns in limb bud cell culture: correlation of in vitro pattern formation with theoretical models.

To experimentally examine whether the pattern changes predicted by theoretical models of pattern formation actually occur in a limb bud cell culture system, we developed a practical method to automatically measure the periodicity of chondrogenic patterns in vitro. The method utilizes binary image processing to quantify the total number of peak and valley pixels in a pattern to obtain the average interval between stripes in the chondrogenic pattern, and we named it the peak length method. The reliability of the peak length method was examined by using computer simulation results. The peak length method enabled us quantitatively obtain the average interval between chondrogenic islands, and the values obtained by this method were closely correlated with the average intervals obtained by manual measurement and two-dimensional Fourier transformation. The average intervals obtained by the peak length method were shown to be stable over a wide range of pattern variations that are frequently observed in actual experiments. By applying the peak length method to actual experimental data, we compared the validity of two theoretical models of pattern formation (cell sorting model and reaction-diffusion model) and it was concluded that the peak-length method is a useful tool to quantitatively analyze chondrogenic patterns in limb micromass culture and to relate theoretical predictions and experimental results of pattern formation.