Abstract
Investigating organ biology requires sophisticated methodologies to induce genetically distinct clones within a tissue. Microscopic analysis of such samples produces information-rich 3D images. However, the 3D nature and spatial anisotropy of clones makes sample analysis challenging and slow and limits the amount of information that can be extracted manually. Here we have developed a pipeline for image processing and statistical data analysis which automatically extracts sophisticated parameters from complex multi-genotype 3D images. The pipeline includes data handling, machine-learning-enabled segmentation, multivariant statistical analysis, and graph generation. This enables researchers to run rigorous analyses on images and videos at scale and in a fraction of the time, without requiring programming skills. We demonstrate the power of this pipeline by applying it to the study of Minute cell competition. We find an unappreciated sexual dimorphism in Minute competition and identify, by statistical regression analysis, tissue parameters that model and predict competitive death.
Competing Interest Statement
The authors have declared no competing interest.
