RootHairML: A Simple Machine-Learning Tool for Quantifying Root Hairs

Root hairs play pivotal roles in nutrient and water acquisition and in plant-microbe interactions. Consequently, understanding the mechanisms underlying root hair development and their regulatory pathways is an important aspect of plant physiology research. Quantifying root hairs and root hair length is often essential in such studies, but is labor-intensive and prone to subjectivity. The availability of straightforward tools for automated root hair measurements is limited, and existing options are often tailored for specific images or do not measure individual root hairs. To address this, we developed RootHairML, a flexible and simple Python-based machine learning tool designed for efficient quantification of root hair lengths. Based on labeled images and pixel features, RootHairML trains a Random Forest model to enable the detection of root hairs in new images. It provides measurements of individual root hair lengths per image and generates annotated images showing all detected root hairs, allowing for manual verification and adjustments. Here, we describe and showcase the use of RootHairML. Overall, RootHairML offers a valuable tool for root hair analysis, enabling researchers to increase data collection and enhance the reproducibility of root hair studies.