High-throughput phenotyping technology for maize roots

This paper describes the development of high-throughput measurement techniques allowing acquisition of phenotypical data describing maize roots. One of a maize root’s traits is the level of complexity, which was expressed in a Fractal Dimension (FD) calculated from root images. Another important trait is the Root Top Angle (RTA) that was measured using a new machine vision algorithm. The measurement system consisted of a semi-automated imaging box that provided a highly diffuse lighting scene and allowing imaging of up to 700 roots per day. The measurement techniques were evaluated using roots recovered from a set of 200 recombinant inbred lines (RILs) derived from a cross between maize inbreds B73 and CML333. B73 and CML333 are known to have different root characteristics and their progeny are expected to show segregation for root traits. Since standard protocols for the measurement of the two root traits are non-existent, no comparisons could be made. Nevertheless, the data showed that the techniques were capable of confirming significant differences in FD among the two inbred lines and their progeny, as well as measuring variations in RTA that are known for the inbreds and their crosses. In addition, first hypotheses about the inheritance of root complexity (as expressed in the FD) and RTA in maize were derived and tested: initial evidence showed that root complexity is a phenotype probably determined by a multitude of genes with small effects. In contrast, the data indicated that the RTA is additively inherited.