High-throughput phenotyping methods for quantifying hair fiber morphology
Quantifying the continuous variation in human scalp hair morphology is of interest to anthropologists, geneticists, dermatologists and forensic scientists, but existing methods for studying hair form are time-consuming and not widely used. Here, we present a high-throughput sample preparation protocol for the imaging of both longitudinal (curvature) and cross-sectional scalp hair morphology. Additionally, we describe and validate a new Python package designed to process longitudinal and cross-sectional hair images, segment them, and provide measurements of interest. Lastly, we apply our methods to an admixed African-European sample (n=140), demonstrating the benefit of quantifying hair morphology over qualitative classification or racial categories, and providing evidence against the long-held belief that cross-sectional morphology predicts curvature.