A statistical algorithm for assessing cellular alignment.

Current statistical techniques for analyzing cellular alignment data in the fields of biomaterials and tissue engineering are limited because of heuristic and less quantitative approaches. For example, generally a cutoff degree limit (commonly 20 degrees) is arbitrarily defined within which cells are considered "aligned." The effectiveness of a patterned biomaterial in guiding the alignment of cells, such as neurons, is often critical to predict relationships between the biomaterial design and biological outcomes, both in vitro and in vivo. This becomes particularly important in the case of peripheral neurons, which require precise axon guidance to obtain successful regenerative outcomes. To address this issue, we have developed a protocol for processing cellular alignment data sets, which implicitly determines an "angle of alignment." This was accomplished as follows: cells "aligning" with an underlying, anisotropic scaffold display uniformly distributed angles up to a cutoff point determined by how effective the biomaterial is in aligning cells. Therefore, this fact was then used to determine where an alignment angle data set diverges from a uniform distribution. This was accomplished by measuring the spacing between the collected, increasingly ordered angles and analyzing their underlying distributions using a normalized cumulative periodogram criterion. The proposed protocol offers a novel way to implicitly define cellular alignment, with respect to various anisotropic biomaterials. This method may also offer an alternative to assess cellular alignment, which could offer improved predictive measures related to biological outcomes. Furthermore, the approach described can be used for a broad range of cell types grown on 2D surfaces, but would not be applicable to 3D scaffold systems in the present format.